Dark | Light
# ![@wey_gu Avatar](https://lunarcrush.com/gi/w:26/cr:twitter::24313912.png) @wey_gu Wey Gu 古思为

Wey Gu 古思为 posts on X about ai, agentic, open ai, code the most. They currently have [------] followers and [---] posts still getting attention that total [---------] engagements in the last [--] hours.

### Engagements: [---------] [#](/creator/twitter::24313912/interactions)
![Engagements Line Chart](https://lunarcrush.com/gi/w:600/cr:twitter::24313912/c:line/m:interactions.svg)

- [--] Week [---------] +29,932%
- [--] Month [---------] +43,446%
- [--] Months [---------] +12,031%
- [--] Year [---------] +13,638%

### Mentions: [--] [#](/creator/twitter::24313912/posts_active)
![Mentions Line Chart](https://lunarcrush.com/gi/w:600/cr:twitter::24313912/c:line/m:posts_active.svg)

- [--] Months [--] +119%
- [--] Year [--] +71%

### Followers: [------] [#](/creator/twitter::24313912/followers)
![Followers Line Chart](https://lunarcrush.com/gi/w:600/cr:twitter::24313912/c:line/m:followers.svg)

- [--] Week [------] +1.10%
- [--] Month [------] +3.50%
- [--] Months [------] +19%
- [--] Year [------] +30%

### CreatorRank: [---------] [#](/creator/twitter::24313912/influencer_rank)
![CreatorRank Line Chart](https://lunarcrush.com/gi/w:600/cr:twitter::24313912/c:line/m:influencer_rank.svg)

### Social Influence

**Social category influence**
[technology brands](/list/technology-brands)  [stocks](/list/stocks)  [social networks](/list/social-networks)  [finance](/list/finance)  [currencies](/list/currencies)  [countries](/list/countries)  [travel destinations](/list/travel-destinations)  [gaming](/list/gaming) 

**Social topic influence**
[ai](/topic/ai), [agentic](/topic/agentic) #548, [open ai](/topic/open-ai), [code](/topic/code), [kimi](/topic/kimi) #314, [claude code](/topic/claude-code) #1138, [openclaw](/topic/openclaw), [6969](/topic/6969), [llm](/topic/llm), [azure](/topic/azure)

**Top accounts mentioned or mentioned by**
[@onlyxuanwo](/creator/undefined) [@nowledgemem](/creator/undefined) [@zhangjintao9020](/creator/undefined) [@thaddeusjiangzh](/creator/undefined) [@frostming90](/creator/undefined) [@yihong0618](/creator/undefined) [@yyhxiaolang](/creator/undefined) [@yetone](/creator/undefined) [@weygu](/creator/undefined) [@leaskh](/creator/undefined) [@strrlthedev](/creator/undefined) [@grok](/creator/undefined) [@danja](/creator/undefined) [@levelsio](/creator/undefined) [@passluo](/creator/undefined) [@elliotchen100](/creator/undefined) [@danbri](/creator/undefined) [@dreamylostcn](/creator/undefined) [@jerryjliu0](/creator/undefined) [@pdmproject](/creator/undefined)

**Top assets mentioned**
[Alphabet Inc Class A (GOOGL)](/topic/$googl) [Fusion (FSN)](/topic/fusion) [StarLink (STARL)](/topic/starlink) [Cloudflare, Inc. (NET)](/topic/cloudflare)
### Top Social Posts
Top posts by engagements in the last [--] hours

"Elixir Redis server http Redis universal  http 100👉 https://docs.statetrace.com/blog/redis-server/ https://docs.statetrace.com/blog/redis-server/"  
[X Link](https://x.com/wey_gu/status/1459195739884568578)  2021-11-12T16:25Z 11.4K followers, [---] engagements


"While this isn't 100% true though I am looking forward to how far it could be very soon #ChatGPT knows #NebulaGraph"  
[X Link](https://x.com/wey_gu/status/1598272058156544000)  2022-12-01T11:05Z 11.3K followers, [--] engagements


"#RaycastApp Pro now"  
[X Link](https://x.com/wey_gu/status/1656214471822704640)  2023-05-10T08:27Z 11.4K followers, [----] engagements


"With This is possible 👍🏻 from langchain.llms import OpenLLM llm = OpenLLM.for_model(model_name='databricks/dolly-v2-3b') llm("What is the difference between a duck and a goose") http://github.com/bentoml/OpenLLM http://github.com/bentoml/OpenLLM"  
[X Link](https://x.com/wey_gu/status/1668433310849318913)  2023-06-13T01:41Z 11K followers, [----] engagements


"@wjsvector @yevazn Yes and the whole infra/stack is open-source too thus you ideally could just fork it and create your own with it. 👉🏻"  
[X Link](https://x.com/wey_gu/status/1678051381389987840)  2023-07-09T16:08Z [----] followers, [--] engagements


"Last week I witnessed Azure OpenAI dropped gpt-3 support(got [---] error) and one of my streamlit online demo was broken. It was interesting that previously azure [---] turbo didnt work perfectly thus I was stick on [--] for long time while in recent version its w/o issues(phew)"  
[X Link](https://x.com/wey_gu/status/1690191948953559041)  2023-08-12T02:42Z [----] followers, [--] engagements


"@appinn What if there is a OpenDAL Desktop/Mobile SIG/SubProject @OnlyXuanwo"  
[X Link](https://x.com/wey_gu/status/1692493746687164427)  2023-08-18T11:08Z [----] followers, [----] engagements


"@danja @danbri Its so very honored seeing OG (linked data) hacker like you trying my demo/NebulaGraph.🫡 Something could be wrong from metad which ended up Auth error was the open file limit changed from [----] please And sorry for the no hosts error too :p"  
[X Link](https://x.com/wey_gu/status/1693043066259554434)  2023-08-19T23:31Z [----] followers, [--] engagements


"@danbri @danja It's OpenAI GPT Quota that ran out"  
[X Link](https://x.com/wey_gu/status/1693223799523856598)  2023-08-20T11:29Z 10.2K followers, [--] engagements


". @Gephi is an amazing GraphViz tool I created #NebulaGraph #Gephi to one click to get gexf file of any query then drop the file into Gephi/Gephi-Lite for graph algo& viz If you are using docker ext its already included just update to try it out"  
[X Link](https://x.com/wey_gu/status/1697782081097719885)  2023-09-02T01:22Z 10.2K followers, [----] engagements


"TIL #NebulaGraph now comes with a decent #scala and #ZIO wrapper project thanks to @dreamylostcn loved that logo"  
[X Link](https://x.com/wey_gu/status/1699379881774293045)  2023-09-06T11:11Z 10.2K followers, [----] engagements


"@danja With docker or Python venv there will be no worries or burden on dependencies of system Python :)"  
[X Link](https://x.com/wey_gu/status/1702808273282637840)  2023-09-15T22:15Z 10.3K followers, [--] engagements


"Excited to delve into Graph #NebulaGraph and Graph RAG with LLM at #DockerCon within the @Docker Community Will you be around in California Join me in LA; let's explore and learn together 👩💻🔍📊 Check out more details: https://bit.ly/dc23speakers https://bit.ly/dc23speakers"  
[X Link](https://x.com/wey_gu/status/1706938692118684123)  2023-09-27T07:47Z 11.3K followers, 28.4K engagements


"This is a great of KG(#NebulaGraph) and LLM experiment summary and tutorial with valuable takeaways Thanks @wenqi_glantz and @jerryjliu0 Was working on things in private last few weeks hopefully I could share/upstream more work soon on this"  
[X Link](https://x.com/wey_gu/status/1707931834385059919)  2023-09-30T01:34Z [----] followers, [----] engagements


"Code Interpreter API and Retrieval API again push down more work to the OpenAI/Cloud which is expected to be the model side(WHY NOT). 🤩"  
[X Link](https://x.com/wey_gu/status/1721447328220160285)  2023-11-06T08:40Z [----] followers, [----] engagements


"@beihuo ASML IaaS case"  
[X Link](https://x.com/wey_gu/status/1729673099962995163)  2023-11-29T01:26Z [----] followers, [----] engagements


"Millions of new materials discovered with Graph deep learning - Google DeepMind #Graph Networks for Materials Exploration (GNoME) deep learning tool dramatically increases the speed and efficiency of discovery by predicting the stability of new materials"  
[X Link](https://x.com/wey_gu/status/1730103331316703726)  2023-11-30T05:55Z [----] followers, [---] engagements


"@genedna 😭"  
[X Link](https://x.com/wey_gu/status/1731121492480708708)  2023-12-03T01:21Z [----] followers, [--] engagements


"It took me quite some hours to make #RedAlert2 run on my #Chromebook smoothly on a 4K screen. I put how-to together in this gist so you don't have to. [--] Chromebook crossover/Windows VM"  
[X Link](https://x.com/wey_gu/status/1741326876474282043)  2023-12-31T05:14Z 10.3K followers, [----] engagements


"Happy new year my friends:)"  
[X Link](https://x.com/anyuser/status/1741821147845738717)  2024-01-01T13:58Z [--] followers, [----] engagements


"@wusheng1108 @ASFSkyWalking @zioscala ping scala hacker @dreamylostcn"  
[X Link](https://x.com/wey_gu/status/1742535564145066202)  2024-01-03T13:17Z [----] followers, [---] engagements


"Python JIT in [----] https://t.co/tsWumVjZDR https://t.co/tsWumVjZDR"  
[X Link](https://x.com/wey_gu/status/1742616896548245928)  2024-01-03T18:40Z 10.2K followers, [--] engagements


"Found in @ayakaneko #NebulaGraph community"  
[X Link](https://x.com/wey_gu/status/1743083102145253790)  2024-01-05T01:32Z 10.3K followers, [----] engagements


"@imwsl90 @bartfastt M1 MacBook pro m1 max dock displaylink cpu dock logo"  
[X Link](https://x.com/wey_gu/status/1743112047775392046)  2024-01-05T03:27Z 10.3K followers, [--] engagements


"@meterscao palm pixi"  
[X Link](https://x.com/anyuser/status/1743161267916034425)  2024-01-05T06:43Z [--] followers, [--] engagements


"🤯 When a "Modernize #NebulaGraph-Python with PDM" issue was created and the next min the creator of @pdm_project @frostming90 just magically showed(together with @yihong0618 ) And [--] min later a PR was shipped with the whole project being polished"  
[X Link](https://x.com/anyuser/status/1743215462358626608)  2024-01-05T10:18Z [--] followers, [----] engagements


"@frostming90 @pdm_project @yihong0618 Aha the time-stamp in the issue thread was twisted so I didn't realize that. Hahaha I like the saying of @yihong0618 's webhook like Spider-Sense of Spider-Man"  
[X Link](https://x.com/anyuser/status/1743220908725301340)  2024-01-05T10:40Z [--] followers, [---] engagements


"@yyhxiaolang 2080ti 22GiB  RTX Titan😭"  
[X Link](https://x.com/anyuser/status/1743286499892740128)  2024-01-05T15:01Z [--] followers, [--] engagements


"@yyhxiaolang openstack cyborg Nvidia vGPU license unlock crack"  
[X Link](https://x.com/wey_gu/status/1743337503502016631)  2024-01-05T18:23Z 10.2K followers, [--] engagements


"@LeaskH founding team bold YouTube clicks iPhone youtuber co-founder"  
[X Link](https://x.com/wey_gu/status/1743462825165779436)  2024-01-06T02:41Z 10.2K followers, [--] engagements


"@LeaskH palm/blackberry pro and con hightlights"  
[X Link](https://x.com/wey_gu/status/1743463979366920343)  2024-01-06T02:46Z 10.2K followers, [--] engagements


"@MiniCssExtract @ChristEd930427 windows mactype mactype"  
[X Link](https://x.com/wey_gu/status/1743465340347224348)  2024-01-06T02:51Z [----] followers, [--] engagements


"@LeaskH  walk through list cmdtab"  
[X Link](https://x.com/wey_gu/status/1743467855444947451)  2024-01-06T03:01Z 10.3K followers, [--] engagements


"@NovaIntrovert"  
[X Link](https://x.com/anyuser/status/1743560956951797890)  2024-01-06T09:11Z [--] followers, [--] engagements


"@yyhxiaolang 👍"  
[X Link](https://x.com/anyuser/status/1743932688871104732)  2024-01-07T09:48Z [--] followers, [---] engagements


"@github @laixintao 🎉"  
[X Link](https://x.com/wey_gu/status/1744028807760404553)  2024-01-07T16:10Z [----] followers, [--] engagements


"@sxyazi Congrats"  
[X Link](https://x.com/wey_gu/status/1744284494461542871)  2024-01-08T09:06Z [----] followers, [---] engagements


"@limbopeng terminal OpenStack flake8 [--] black [--] max line 100"  
[X Link](https://x.com/wey_gu/status/1747571375324176824)  2024-01-17T10:47Z 10.3K followers, [--] engagements


"@thecalicastle @yyhxiaolang :-D VueJS"  
[X Link](https://x.com/wey_gu/status/1750412520567734329)  2024-01-25T06:57Z 10.2K followers, [--] engagements


"@yyhxiaolang cc @thecalicastle 🥹 useEffect VueJS 🤣"  
[X Link](https://x.com/wey_gu/status/1751922475363967026)  2024-01-29T10:57Z 10.3K followers, [--] engagements


"@sxyazi congrats"  
[X Link](https://x.com/wey_gu/status/1752999184083386522)  2024-02-01T10:15Z 10.3K followers, [--] engagements


"I can't wait to see Apple Vision Pro become the size of Google Glass one day"  
[X Link](https://x.com/wey_gu/status/1753951317368750544)  2024-02-04T01:19Z 10.3K followers, [----] engagements


"@yihong0618 Patpat - OpenAI Azure API IP iAP apple id plus - Colab NebulaGraph-Lite block ** Colab 😭"  
[X Link](https://x.com/wey_gu/status/1765280723827638678)  2024-03-06T07:38Z 10.4K followers, [--] engagements


"@zhangjintao9020 @yihong0618 Google Play Plus session 😢 plus API"  
[X Link](https://x.com/wey_gu/status/1765320613701263788)  2024-03-06T10:16Z 10.4K followers, [--] engagements


"@beautyyuyanli 🥳 team #NebulaGraph GenAI will physically union next week"  
[X Link](https://x.com/wey_gu/status/1770784402567528511)  2024-03-21T12:07Z 10.4K followers, [--] engagements


"scala java  guobin scala hacker🫡"  
[X Link](https://x.com/wey_gu/status/1775385301021814854)  2024-04-03T04:50Z 10.4K followers, [--] engagements


"@oran_ge Graph RAG LlamaIndex workshop - - - - https://bit.ly/graph-rag-workshop https://www.siwei.io/demo-dumps/local-llm/Graph_RAG_Local.html https://www.siwei.io/demo-dumps/graph-rag/GraphRAG.html https://www.siwei.io/tutors/GraphRAG/101.html https://bit.ly/graph-rag-workshop https://www.siwei.io/demo-dumps/local-llm/Graph_RAG_Local.html https://www.siwei.io/demo-dumps/graph-rag/GraphRAG.html https://www.siwei.io/tutors/GraphRAG/101.html"  
[X Link](https://x.com/wey_gu/status/1785107395938804031)  2024-04-30T00:42Z 10.5K followers, [----] engagements


"Check out my X card from moimoi @moimoi_app 🎉 #moimoi #digitalmosaic"  
[X Link](https://x.com/wey_gu/status/1785289630440387070)  2024-04-30T12:46Z 10.5K followers, [----] engagements


"Graph RAG with GNN. A framework for evaluating Retrieval-Augmented Text Generation for #LLMs & Graph #RAG grows in complexity It can be organized paradigm into categories SURGE uses #GNN to extract subgraphs from #knowledgegraph achieving customization #EmergingTech #AI https://t.co/Nj0Zhg3N2d https://t.co/cmy4zWpsR9 A framework for evaluating Retrieval-Augmented Text Generation for #LLMs & Graph #RAG grows in complexity It can be organized paradigm into categories SURGE uses #GNN to extract subgraphs from #knowledgegraph achieving customization #EmergingTech #AI https://t.co/Nj0Zhg3N2d"  
[X Link](https://x.com/wey_gu/status/1785311399293419932)  2024-04-30T14:13Z 10.5K followers, [----] engagements


"Some of my old Graph RAG stuff is worth a retweet"  
[X Link](https://x.com/wey_gu/status/1785312891224174975)  2024-04-30T14:18Z 10.5K followers, [--] engagements


"🧠 Access Distributed Graph Data(#NebulaGraph) in the great OpenDAL manner This will be sick😱. Thanks @OnlyXuanwo and Su Yan Gege for making this possible A joint collaboration & opportunity OPFS Apache OpenDAL and @NebulaGraph Hello everyone. Here's an announcement from the #OpenDAL community regarding participation in the Open Source Promotion event. This year we are featuring three projects: OPFS monoio and @NebulaGraph's support Feel free to visit https://t.co/U1i7EThdMC for more details. Hello everyone. Here's an announcement from the #OpenDAL community regarding participation in the"  
[X Link](https://x.com/wey_gu/status/1785324096034586647)  2024-04-30T15:03Z 10.5K followers, [----] engagements


"@JustinLin610 Thanks"  
[X Link](https://x.com/wey_gu/status/1785567271072715106)  2024-05-01T07:09Z 10.5K followers, [---] engagements


"@9hills LiteLLM Azure OAI OAI Function Calling"  
[X Link](https://x.com/wey_gu/status/1787679736690352176)  2024-05-07T03:03Z 10.5K followers, [--] engagements


"Dannys work makes dev on macOS whole next level comparing to wsl2/docker desktop on windows. https://github.com/microsoft/WSL/issues/9555 In the next version of OrbStack: true near-native file system performance 210x faster than before pnpm: 0.5% overhead vs. native macOS (avg [----] vs [-----] sec) with [----] packages https://t.co/wSQIUMvbLQ https://github.com/microsoft/WSL/issues/9555 In the next version of OrbStack: true near-native file system performance 210x faster than before pnpm: 0.5% overhead vs. native macOS (avg [----] vs [-----] sec) with [----] packages https://t.co/wSQIUMvbLQ"  
[X Link](https://x.com/wey_gu/status/1789974950901608553)  2024-05-13T11:04Z 10.6K followers, [----] engagements


"chatgpt-4o streaming glass"  
[X Link](https://x.com/wey_gu/status/1790591847557149148)  2024-05-15T03:55Z 10.6K followers, [--] engagements


"@strrlthedev hypervisor vsphereesxi"  
[X Link](https://x.com/wey_gu/status/1792781889457324087)  2024-05-21T04:58Z 10.6K followers, [--] engagements


"@strrlthedev track license ESXi vsphere"  
[X Link](https://x.com/wey_gu/status/1792787630494797922)  2024-05-21T05:20Z 10.6K followers, [--] engagements


"options - raycast ad-hoc cli helper raycast ai - GitHub copilot cli - cusor terminal commad+k"  
[X Link](https://x.com/wey_gu/status/1792791599161966866)  2024-05-21T05:36Z 10.6K followers, [--] engagements


"@yhcnux warpOpt + Click Tmux select Terminal"  
[X Link](https://x.com/wey_gu/status/1792804319521374320)  2024-05-21T06:27Z 10.6K followers, [--] engagements


"#NebulaGraph + Apache DAL nodejs client  ArkTS Palm WebOS"  
[X Link](https://x.com/wey_gu/status/1793251043582628072)  2024-05-22T12:02Z 10.6K followers, [--] engagements


"@TheYotg Thanks for sharing this The GitHub page url is [---] though. https://mlx-graphs.github.io/mlx-graphs/ https://mlx-graphs.github.io/mlx-graphs/"  
[X Link](https://x.com/wey_gu/status/1798732848826073581)  2024-06-06T15:05Z 10.7K followers, [---] engagements


"@faouzia You the best in this performance. As one big fan of Justin Timberlake its sooo enjoyable listening to your masterpiece version"  
[X Link](https://x.com/wey_gu/status/1801677810706301217)  2024-06-14T18:07Z 10.7K followers, [---] engagements


"I was always planning to leverage the amazing @textualizeio to create a terminal #NebulaGraph UI tonight I had some time to see @tedconbeer 's great the SQL IDE now to be a Database IDE after supporting @NebulaGraph). $ pip install harlequin-nebulagraph $ harlequin -a nebulagraph demo video in the thread. https://github.com/wey-gu/harlequin-nebulagraph http://harlequin.sh http://harlequin.sh https://github.com/wey-gu/harlequin-nebulagraph http://harlequin.sh http://harlequin.sh"  
[X Link](https://x.com/wey_gu/status/1802708227014213989)  2024-06-17T14:21Z 10.7K followers, [----] engagements


"Azure OpenAI(Azure OAI) #NebulaGraph team  Azure OAI serving/client OAI  tricky OAI function calling LiteLLM OAI function calling"  
[X Link](https://x.com/wey_gu/status/1805479884535259574)  2024-06-25T05:55Z 10.7K followers, [--] engagements


"@xsser_w IP IP Reputation 😭 OpenAI traffic https call"  
[X Link](https://x.com/wey_gu/status/1805558896049770814)  2024-06-25T11:09Z 10.7K followers, [--] engagements


"psw3c vue principle  tailwind css v0/shadcn empower 🥹"  
[X Link](https://x.com/wey_gu/status/1809752226103193891)  2024-07-07T00:52Z 10.8K followers, [--] engagements


"A great milestone. LlamaCloud is solving the key problem s elegantly for GenAI workflows. Or I should say knowledge/data driven workflow. Im excited to announce a beta release of LlamaCloud 🦙⛅ - the data processing layer for your LLM applications. Weve opened up a waitlist for users to signup To build a better knowledge assistant you need an advanced data processing layer to prevent the garbage in = garbage https://t.co/zk43Tg3CgO Im excited to announce a beta release of LlamaCloud 🦙⛅ - the data processing layer for your LLM applications. Weve opened up a waitlist for users to signup To"  
[X Link](https://x.com/wey_gu/status/1810723105733489080)  2024-07-09T17:10Z 10.8K followers, [----] engagements


"@vanillaCitron 🙏"  
[X Link](https://x.com/wey_gu/status/1813512295173275938)  2024-07-17T09:53Z 10.8K followers, [---] engagements


"The abstraction of LlamaIndex is data/knowledge-driven based on which we could build GenAI apps with ease and peace. The way that I use/learn LlamaIndex is to walkthrough both docs and code(of course by contributing to it) While for some who enjoyed reading in a real book there is one for you😊 and now I am thinking its a good start for rag-101 too(I used to suggest new dev to walkthrough LI docs to get started on RAG) I liked how it was structured in the book. Thanks @Vinishkakalra for sending me the preview copy."  
[X Link](https://x.com/wey_gu/status/1814258988768203124)  2024-07-19T11:20Z 10.9K followers, [----] engagements


"WE ARE HIRING @beautyyuyanli #NebulaGraph GenAI - build Graph + GenAI Magic in public - scale the magic to other wizards/witches to help them do the same - hack because we can - change the world with #NebulaGraph #ISOGQL and #OpenSource Join us We are seeking hackers based in Shanghai: - Frontend hacker - Backend hacker(Python) Reach out to me or our recruiter @eva_wuwuwu please"  
[X Link](https://x.com/wey_gu/status/1814267686563566019)  2024-07-19T11:54Z 10.9K followers, 27.9K engagements


"@OnlyXuanwo happy birthday xuanwo"  
[X Link](https://x.com/wey_gu/status/1815126625165455426)  2024-07-21T20:48Z 10.9K followers, [---] engagements


"- copilot ++ - block selection change/apply - chat sota - sonnet gpt-4 chat apply - composer tier cursor"  
[X Link](https://x.com/wey_gu/status/1820678802646978674)  2024-08-06T04:30Z 10.9K followers, [--] engagements


"@kelseyhightower highly recommend to try orbstack as replacement of both VMware fusion and docker desktop"  
[X Link](https://x.com/wey_gu/status/1825344770199167328)  2024-08-19T01:31Z 11.3K followers, [----] engagements


"I have an intel PixelBook with quite good battery life apart from all my other Mac machines its perfectly designed and performs good enough for some ad-hoc dev work. The only thing I struggled with is my muscle memory of key mappings. I used to use Ubuntu desktop as daily driver for years when electron-based apps were not that popular as its today but only quite few use cases struggled already(launcher like raycast/alfred meeting app for screen sharing)"  
[X Link](https://x.com/wey_gu/status/1825614550596497465)  2024-08-19T19:23Z 11.3K followers, [---] engagements


"Were thrilled to announce that @repsiace is joining the #NebulaGraph GenAI team Excited to see the amazing things well build together. Welcome aboard 🚀 🚀 Excited to start my journey at #NebulaGraph today transitioning from Rust to Python for GenAI 🚀 Excited to start my journey at #NebulaGraph today transitioning from Rust to Python for GenAI"  
[X Link](https://x.com/wey_gu/status/1833302691038249473)  2024-09-10T00:33Z 11.3K followers, [----] engagements


"metas xr glass seems to be in shape I dreamed to have in google glass"  
[X Link](https://x.com/wey_gu/status/1839285671501959356)  2024-09-26T12:47Z 11.3K followers, [---] engagements


"Along with Yanli and PsiACE we're so very happy to welcome to the NebulaGraph GenAI Team 😄 https://github.com/boyyangzai https://github.com/boyyangzai"  
[X Link](https://x.com/wey_gu/status/1844257974631661611)  2024-10-10T06:05Z 11.2K followers, [----] engagements


"@yuyinws @nuxt_hub @stackblitz Great work Leo If you are interested in persisting pkg-graph in Graph Database to enable more graph query/ graph insight we are happy to sponsor you via #NebulaGraph Cloud"  
[X Link](https://x.com/wey_gu/status/1844944050652082677)  2024-10-12T03:31Z 11.2K followers, [---] engagements


"@yuyinws @nuxt_hub @stackblitz I meant other than getting a subgraph from one node from the whole graph compute graph algorithms and persist features back to the graph and visualize based on graph algo(node importance community detection etc.)"  
[X Link](https://x.com/wey_gu/status/1844944998061506654)  2024-10-12T03:35Z 11.2K followers, [---] engagements


"Now I can code (w/ cursor) even on a plane w/o starlink🤣. Qwen [---] Coder 32B on mlx_community thanks to super @Prince_Canuma Thanks 🙏 Let me try it now https://t.co/r5fINv611s Qwen [---] Coder 32B on mlx_community thanks to super @Prince_Canuma Thanks 🙏 Let me try it now https://t.co/r5fINv611s"  
[X Link](https://x.com/wey_gu/status/1856330413012078739)  2024-11-12T13:37Z 11.3K followers, [----] engagements


"@zhangjintao9020 @thekonginc"  
[X Link](https://x.com/wey_gu/status/1857363347005456463)  2024-11-15T10:01Z 11.6K followers, [---] engagements


"@Akinsanmidev @buaaxhm drop your question here or in a github issue/stackoverflow post"  
[X Link](https://x.com/wey_gu/status/1859424199162593603)  2024-11-21T02:30Z 11.4K followers, [--] engagements


"So very exciting for Qwen2-VL support and MLX. Coming this month: a complete rewrite of the Ollama engine. Includes CUDA support and more models. Coming in 2025: MLX support https://t.co/PcdA1xtxvQ Coming this month: a complete rewrite of the Ollama engine. Includes CUDA support and more models. Coming in 2025: MLX support https://t.co/PcdA1xtxvQ"  
[X Link](https://x.com/wey_gu/status/1863788988034867334)  2024-12-03T03:34Z 11.4K followers, [---] engagements


"GitHub [----] 😋"  
[X Link](https://x.com/wey_gu/status/1868994904917262486)  2024-12-17T12:21Z 11.6K followers, [----] engagements


"2024 Raycast wrapped"  
[X Link](https://x.com/wey_gu/status/1869202554368012783)  2024-12-18T02:06Z 11.6K followers, [---] engagements


"o1 preview was slow yet truly valuable. o1 is already fast deepseek r1 and qwen qwq are open-weight Gemini thinking is reasoning better yet visible. Now we have o3 and o3-mini in new level of reasoning. Reasoning models to be norm as agentic backbone is soon to come🥳 Day 12: Early evals for OpenAI o3 (yes we skipped a number) https://t.co/iWXg9IGuZM Day 12: Early evals for OpenAI o3 (yes we skipped a number) https://t.co/iWXg9IGuZM"  
[X Link](https://x.com/wey_gu/status/1870246578268844521)  2024-12-20T23:15Z 11.6K followers, [---] engagements


"@OikawaRizumu @vuejsnation"  
[X Link](https://x.com/wey_gu/status/1871506424871985623)  2024-12-24T10:41Z 11.6K followers, [---] engagements


"Wow thanks qwen team merry Christmas 🎄 Happy holidays This is something fun for you to play with. QVQ the new model can think and reason on images 🎄 Happy holidays This is something fun for you to play with. QVQ the new model can think and reason on images"  
[X Link](https://x.com/wey_gu/status/1871608489350685124)  2024-12-24T17:26Z 11.6K followers, [---] engagements


"notebookLM 🤯 I created an open-source implementation of @notebooklm 🤖 Deepseek-V3 API using @OpenRouterAI 🎙 @play_ht TTS using @FAL API 📚 Create AI podcasts on ANY topic 100% Customizable All this in [--] lines of code Check out the GitHub repo: https://t.co/Jp96O5aBci https://t.co/ivqk8rYNaS 🤯 I created an open-source implementation of @notebooklm 🤖 Deepseek-V3 API using @OpenRouterAI 🎙 @play_ht TTS using @FAL API 📚 Create AI podcasts on ANY topic 100% Customizable All this in [--] lines of code Check out the GitHub repo: https://t.co/Jp96O5aBci https://t.co/ivqk8rYNaS"  
[X Link](https://x.com/wey_gu/status/1875480877490385065)  2025-01-04T09:54Z 11.6K followers, [---] engagements


"The GenAI-based content creation engine refly is source-available today a great call on this:) https://github.com/refly-ai/refly https://github.com/refly-ai/refly"  
[X Link](https://x.com/wey_gu/status/1882731935245758869)  2025-01-24T10:07Z 11.7K followers, [---] engagements


"@jerryjliu0 @llama_index @databricks @KPMG congrats"  
[X Link](https://x.com/wey_gu/status/1918073235775865231)  2025-05-01T22:41Z 12K followers, [--] engagements


"Interesting lexical patterns analysis based clustering showing the distilling lineages between latest models. R1-0528 showed closer to Gemini while R1 to OpenAI. DeepSeek-R1-0528 Gemini R1 OpenAI DeepSeek R1 Google Gemini R1 OpenAI llama https://t.co/Qr1L5c5Jid DeepSeek-R1-0528 Gemini R1 OpenAI DeepSeek R1 Google Gemini R1 OpenAI llama https://t.co/Qr1L5c5Jid"  
[X Link](https://x.com/wey_gu/status/1928625698438045897)  2025-05-31T01:32Z 12.1K followers, [---] engagements


"source code is the new target in GenAI era. design docs meta source code and their change histories are true sources now  AI Coding  AI Coding LLM AI Coding  AI Coding  AI Coding LLM AI Coding"  
[X Link](https://x.com/wey_gu/status/1930663917187149921)  2025-06-05T16:32Z 12.2K followers, [---] engagements


"Cua introduced windows sandbox support🫡 computer = Computer( provider_type=VMProviderType.WINSANDBOX os_type="windows" memory="4GB" ) Introducing Windows Sandbox support - run computer-use agents on Windows business apps without VMs or cloud costs. 1/6 https://t.co/uL6LOMifAs Introducing Windows Sandbox support - run computer-use agents on Windows business apps without VMs or cloud costs. 1/6 https://t.co/uL6LOMifAs"  
[X Link](https://x.com/wey_gu/status/1935754172231827548)  2025-06-19T17:38Z 12.3K followers, [---] engagements


"@frostming90 @dinah_zhang @Piglei"  
[X Link](https://x.com/wey_gu/status/1937703696215024060)  2025-06-25T02:45Z 12.3K followers, [--] engagements


"This work showed RL transfers SFT barely does. RL generalizes SFT doesn't. Self-exploration and feedback leads to AGI not SFT. Does that mean once we reach certain intelligence level (able to generate valid candidate) the value of human generated data is diminishing. And feedback especially rule generated massive RL generalizes SFT doesn't. Self-exploration and feedback leads to AGI not SFT. Does that mean once we reach certain intelligence level (able to generate valid candidate) the value of human generated data is diminishing. And feedback especially rule generated massive"  
[X Link](https://x.com/wey_gu/status/1940600476221006188)  2025-07-03T02:36Z 12.3K followers, [---] engagements


"@lantianlaoli openai-translator"  
[X Link](https://x.com/wey_gu/status/1941574503433371923)  2025-07-05T19:06Z 12.4K followers, [---] engagements


"@zhangjintao9020 @himself_65 @strrlthedev dxt client eval claude desktop/ cursor/ chatwise client shared sandbox nodejs runtime"  
[X Link](https://x.com/wey_gu/status/1945073440018338243)  2025-07-15T10:50Z 12.4K followers, [---] engagements


"chatwise + groq k2 + websearch(llama on groq) ad-hoc search k2 on groq would be a killing existence for certain usage k2 on groq would be a killing existence for certain usage"  
[X Link](https://x.com/wey_gu/status/1945076640809160805)  2025-07-15T11:03Z 12.4K followers, [----] engagements


"Still more works to do #buildinpublic 😭 #buildinpublic https://t.co/sBdSkIS0FV 😭 #buildinpublic https://t.co/sBdSkIS0FV"  
[X Link](https://x.com/wey_gu/status/1946030985596043715)  2025-07-18T02:15Z 12.5K followers, [----] engagements


"@zhangjintao9020 Sort of But not to touch the Note-App area just help sink to/explore from our key insights threads across Precious TALKS with GenAI agents"  
[X Link](https://x.com/wey_gu/status/1946036137812021600)  2025-07-18T02:35Z 12.4K followers, [---] engagements


"Hey @grok Who are my potential ones to follow from my 2-hop top mutuals but not yet connected yet with my most shared top mutuals"  
[X Link](https://x.com/wey_gu/status/1946390650557243485)  2025-07-19T02:04Z 12.5K followers, [---] engagements


"@grok @yetone @OnlyXuanwo @oran_ge @YannLecun @lexfridman @ylecun @drfeifei Yeah so I know you are lacking of api expressions to actually do such query right"  
[X Link](https://x.com/wey_gu/status/1946392498596897056)  2025-07-19T02:11Z 12.5K followers, [--] engagements


"@grok @yetone @OnlyXuanwo @YannLecun @lexfridman @ylecun @drfeifei @karpathy Thanks for the transparency"  
[X Link](https://x.com/wey_gu/status/1946396066829787355)  2025-07-19T02:25Z 12.5K followers, [--] engagements


"@grok @OikawaRizumu @QuantumTransf @Nickel_Angel @shiratsuyudachi @Meowneko_Mio @JakkuSakura @Eureka222323 @kiligulyanov @RobertoC_cr03x @misaka__network @saonanaa @kz152cx @sCuSO_4 @ggccsxhn @Exusiai159 @isaac_eins @aqCuSO_4 @InfectQWQ @ChinTAT_2008 @vanto156607 @AFeiSigniFy__ @bghtnya @yqua_ @bushi_8964 @sxshuishuimao @qingshixiang258 @hiner233 @Chiyuyu520 @Miisago_han @xiaoyieQAQ @VicoOci @mitsuki87613 @yunliu2005 @All12240519 @jiangkedou @ilkos_sdvx @Nickel_Schwarz @dimangmu8964 @89JinPing64 @DDzhanshou5150 @FoxRed695449 @Cyan_0731 @lcMenci @ProbiusOfficial @mike_cheng1208 @Stv_Lynn"  
[X Link](https://x.com/wey_gu/status/1946433389114143209)  2025-07-19T04:54Z 12.5K followers, [----] engagements


"for me it's still cursor. my workflow is more pair coding with cursor and pair/vibe design with v0 dot dev. in study phase i talked to gemini pro/o4 mini/claude over my upstream projects within github copilot; run deep research over chatgpt/gemini; then i started writing RFC within the repo back and forth the study and poc code with cursor's agent to finalize feasibility and arch design etc. then i'll have main arch/feature design and i ask cursor to sync things during my iteration of design. and sinking/updating cursor rules from those design docs(keeps sharp and short). then i started to"  
[X Link](https://x.com/wey_gu/status/1946497903020884282)  2025-07-19T09:10Z 12.9K followers, 15.4K engagements


"For me its more pair coding rather than vibe coding now when building things. @yetone for me it's still cursor. my workflow is more pair coding with cursor and pair/vibe design with v0 dot dev. in study phase i talked to gemini pro/o4 mini/claude over my upstream projects within github copilot; run deep research over chatgpt/gemini; then i started writing RFC @yetone for me it's still cursor. my workflow is more pair coding with cursor and pair/vibe design with v0 dot dev. in study phase i talked to gemini pro/o4 mini/claude over my upstream projects within github copilot; run deep research"  
[X Link](https://x.com/wey_gu/status/1946550532178620522)  2025-07-19T12:39Z 12.5K followers, [---] engagements


"@suavesu Omg didnt know the license switch from gplv3 to apache2"  
[X Link](https://x.com/wey_gu/status/1946822260180005253)  2025-07-20T06:39Z 12.5K followers, [---] engagements


"Now I am running @nowledgelabs at night and my free time only.(alpha launch soon) Hopefully will make it my daytime soon Still more works to do #buildinpublic https://t.co/dgkB5KXWCF Still more works to do #buildinpublic https://t.co/dgkB5KXWCF"  
[X Link](https://x.com/wey_gu/status/1947609678386761867)  2025-07-22T10:48Z 12.5K followers, [----] engagements


"original precision [--] gib ram 🫡 You can now run gpt-oss-120b & 20b locally with our GGUFs 🦥 Run OpenAI's 120b model on 66GB RAM & 20b model on 14GB RAM. Both in original precision. Uploads includes our chat template fixes. Guide: https://t.co/LZPNUvyy2G GGUF: https://t.co/mfZDpekRER https://t.co/VUmkjjs5ZX You can now run gpt-oss-120b & 20b locally with our GGUFs 🦥 Run OpenAI's 120b model on 66GB RAM & 20b model on 14GB RAM. Both in original precision. Uploads includes our chat template fixes. Guide: https://t.co/LZPNUvyy2G GGUF: https://t.co/mfZDpekRER https://t.co/VUmkjjs5ZX"  
[X Link](https://x.com/wey_gu/status/1952904749457784916)  2025-08-06T01:29Z 12.7K followers, [---] engagements


"We need an universal well-known paths to cache those LLM/Embedding models on-device"  
[X Link](https://x.com/wey_gu/status/1955533922433167496)  2025-08-13T07:36Z 12.7K followers, [----] engagements


"@frostming90 Ollama is heavy yet not support mlx/onnx yet for now"  
[X Link](https://x.com/wey_gu/status/1955554561504244136)  2025-08-13T08:58Z 12.7K followers, [---] engagements


"@OnlyXuanwo @tison1096 +1 for dingtalk and huawei im"  
[X Link](https://x.com/wey_gu/status/1955620149245181966)  2025-08-13T13:19Z 12.7K followers, [---] engagements


"@hylarucoder gpt5 troubleshooting"  
[X Link](https://x.com/wey_gu/status/1956259419035197572)  2025-08-15T07:39Z 12.7K followers, [--] engagements


"@repsiace we are chain of wechat messages"  
[X Link](https://x.com/wey_gu/status/1956274447528419442)  2025-08-15T08:39Z 12.7K followers, [--] engagements


"@localhost_4173 ping @JustinLin610"  
[X Link](https://x.com/wey_gu/status/1958113259145531711)  2025-08-20T10:25Z 12.8K followers, [---] engagements


"@ZinkLu2 @nowledgelabs remote llm 🫡🧎"  
[X Link](https://x.com/wey_gu/status/1958382322191151157)  2025-08-21T04:15Z 12.8K followers, [---] engagements


"we could define the term vibe here further to me i do pair coding rather than vibe coding while pair coding with genai is a just the form of broader definition of vide coding in most context . vibe coding  . vibe coding"  
[X Link](https://x.com/wey_gu/status/1958484591335133538)  2025-08-21T11:01Z 12.8K followers, [---] engagements


"https://github.com/deepseek-ai/DeepSeek-V3/issues/849 https://github.com/deepseek-ai/DeepSeek-V3/issues/849"  
[X Link](https://x.com/wey_gu/status/1960486748305371364)  2025-08-26T23:37Z 12.9K followers, [--] engagements


"semtools with claude code Turn Claude Code into a Financial Analyst 🤖💹 In this video we point Claude Code at a bucket of 10k filing PDFs and have it perform complex analysis across the entire set of docs Claude Code doesnt have file understanding out of the box (it kind of does but its terrible / https://t.co/N5EuyJw5PS Turn Claude Code into a Financial Analyst 🤖💹 In this video we point Claude Code at a bucket of 10k filing PDFs and have it perform complex analysis across the entire set of docs Claude Code doesnt have file understanding out of the box (it kind of does but its terrible /"  
[X Link](https://x.com/wey_gu/status/1962614639344148507)  2025-09-01T20:32Z 12.9K followers, [----] engagements


"@hd_nvim This was in our mind too while just didnt have chance to actually impl. Define proper graph schema code parsing mechanism and finally agentic operators to leverage the graph should help and ideally it could scales yet latency wise to be performant"  
[X Link](https://x.com/wey_gu/status/1964220953736728702)  2025-09-06T06:55Z 13K followers, [---] engagements


"SGLang mooncake integration out of box support We are excited to announce SGLang HiCache our community solution for hierarchical KV caching to power high-performance LLM serving. ⚡ Performance: up to [--] throughput and 80% TTFT reduction demonstrated in benchmarks and real-world deployments. 🗂 Flexibility: seamless https://t.co/9pPjFbuGuT We are excited to announce SGLang HiCache our community solution for hierarchical KV caching to power high-performance LLM serving. ⚡ Performance: up to [--] throughput and 80% TTFT reduction demonstrated in benchmarks and real-world deployments. 🗂 Flexibility:"  
[X Link](https://x.com/wey_gu/status/1966442241192337479)  2025-09-12T10:02Z 13K followers, [----] engagements


"@shinyzhu cloudflare"  
[X Link](https://x.com/wey_gu/status/1969230746037461492)  2025-09-20T02:42Z 13K followers, [---] engagements


"@Kechen_Li2018 nuxtjs remix svelte I am not with fe background actually"  
[X Link](https://x.com/wey_gu/status/1970859874113536047)  2025-09-24T14:36Z 13.1K followers, [--] engagements


"TIL Brahmand an opensource stateless graph query/computation layer on ClickHouse that speaks OpenCypher. https://github.com/darshanDevrai/brahmand Introducing Brahmand: an open source Graph Engine built on top of ClickHouse Extending ClickHouse with graph modeling and OpenCypher merging OLAP speed with graph analysis. Brahmand is a high-performance stateless graph-analysis layer for ClickHouse. Brahmand is targeting https://t.co/JS3RQzgXsB https://github.com/darshanDevrai/brahmand Introducing Brahmand: an open source Graph Engine built on top of ClickHouse Extending ClickHouse with graph"  
[X Link](https://x.com/wey_gu/status/1971018073445630277)  2025-09-25T01:05Z 13.1K followers, [----] engagements


"Kwais coding model looks amazingly good in its size. 🚀Introducing KAT-Dev-32B & KAT-Coder -KAT-Dev-32B62.4% on SWE-Bench Verified top-5 among open-source models -KAT-Coder73.4% on SWE-Bench Verified rivals top proprietary models 👇 Try it 🤗 Model: https://t.co/IF20Ptd8FF 🔗 Blog: https://t.co/021ZrW7caQ https://t.co/U8mYXkqpiV 🚀Introducing KAT-Dev-32B & KAT-Coder -KAT-Dev-32B62.4% on SWE-Bench Verified top-5 among open-source models -KAT-Coder73.4% on SWE-Bench Verified rivals top proprietary models 👇 Try it 🤗 Model: https://t.co/IF20Ptd8FF 🔗 Blog: https://t.co/021ZrW7caQ"  
[X Link](https://x.com/wey_gu/status/1972461359733629236)  2025-09-29T00:40Z 13.1K followers, [----] engagements


"Apache Gravitino makes datalake wirable to agentic stacks 👍 🚀 Apache Gravitino 1.0.0 is here Metadata Goes Agentic. From Metadata Management to Contextual Engineering: ⚡ Stats Policy Job systems 🤖 MCP server (agent-ready) 🔒 RBAC governance 🔗 https://t.co/DyOcGHB4tT #OpenSource #AI #Metadata #Datastrato #ApacheGravitino https://t.co/6wUsgekiVe 🚀 Apache Gravitino 1.0.0 is here Metadata Goes Agentic. From Metadata Management to Contextual Engineering: ⚡ Stats Policy Job systems 🤖 MCP server (agent-ready) 🔒 RBAC governance 🔗 https://t.co/DyOcGHB4tT #OpenSource #AI #Metadata #Datastrato"  
[X Link](https://x.com/wey_gu/status/1973535247242502646)  2025-10-01T23:47Z 13.1K followers, [---] engagements


"Thanks for the shoutout @chatmcp Thrilled to integrate Nowledge Mem (impressed on how seamlessly it was done and the great UX of into the MCP ecosystemempowering local agents with graph-augmented memory To celebrate use promo code MCPSO95 for an almost-free upgrade to Nowledge Mem Pro (limited time) Demo: Join alpha waitlist: Intro blog: Let's build smarter knowledge flywheels together #AI #KnowledgeGraph #MCP https://nowled.ge/blog-intro-mem https://nowled.ge/alpha https://x.com/nowledgelabs/status/1952578790775017977 http://mcp.so http://mcp.so Great to have Nowledge Mem MCP on board"  
[X Link](https://x.com/wey_gu/status/1973782081709965352)  2025-10-02T16:08Z 13.2K followers, 13K engagements


"@hd_nvim @OpenAI"  
[X Link](https://x.com/wey_gu/status/1975494910179770664)  2025-10-07T09:34Z 13.2K followers, [---] engagements


"Forgetting is a gift. ChatGPT can now automatically manage your saved memoriesno more memory full. You can also search and sort memories by recency and choose which to re-prioritize in settings. Rolling out to Plus and Pro users on the web globally starting today. https://t.co/T1vSNH5289 https://t.co/xRHLFTu2Am ChatGPT can now automatically manage your saved memoriesno more memory full. You can also search and sort memories by recency and choose which to re-prioritize in settings. Rolling out to Plus and Pro users on the web globally starting today. https://t.co/T1vSNH5289"  
[X Link](https://x.com/wey_gu/status/1978640008124018772)  2025-10-16T01:51Z 13.3K followers, [---] engagements


"a k8s native approach is on the way. Both OpenEdx and tutor community are quite active and friendly👍 I will for sure explore more And I am happy to have a chance to contribute to the community:"  
[X Link](https://x.com/wey_gu/status/1388331359274377224)  2021-05-01T03:16Z 13.8K followers, [--] engagements


"1. Fixed some invalid commands inthe openedx native deployment documentation [--]. Fixed LMS uwsgi [---] error when request header size exceeds 4096K PR: https://github.com/overhangio/tutor/pull/427 https://openedx.atlassian.net/wiki/spaces/OpenOPS/pages/60227913/Managing+Open+edX+Tips+and+Tricks https://github.com/overhangio/tutor/pull/427 https://openedx.atlassian.net/wiki/spaces/OpenOPS/pages/60227913/Managing+Open+edX+Tips+and+Tricks"  
[X Link](https://x.com/wey_gu/status/1388332495486156800)  2021-05-01T03:20Z 13.8K followers, [--] engagements


"@strrlthedev opps just go to model providers to know this too 🤣"  
[X Link](https://x.com/wey_gu/status/1928297580296896573)  2025-05-30T03:49Z 14.7K followers, [---] engagements


"Cursor is one of the few implements prompts rather than only tools of the MCP protocol. Thus in cursor you can call @NowledgeMem in the chat to summarize then memorize your key insights in the thread. And all we need is to type / then mem itll focus on prompt from Nowledge Mem MCP and enter to call it. Now cursor allows us to map this flow to ANY prompt in user space. make your own commands /fix /refactor /plan soon in @cursor_ai ☺ make your own commands /fix /refactor /plan soon in @cursor_ai ☺"  
[X Link](https://x.com/wey_gu/status/1964005177960362251)  2025-09-05T16:38Z 13.8K followers, [----] engagements


". @NowledgeMem was built to bring us the freedom of Agents/AI Tools build our own context flywheel😉 OpenAI Agent SDK OpenAI API Demo IP OpenAI  ChatGPT OpenAI Agent SDK OpenAI API Demo IP OpenAI  ChatGPT"  
[X Link](https://x.com/wey_gu/status/1978719936601104789)  2025-10-16T07:09Z 13.3K followers, [----] engagements


"Upgraded from NextJS [--] to 16.0.0: - local dev build is way better(not that good as vite though) - production build is way better(4min30sec to 1min28sec) Next.js [--] Cache Components Turbopack enabled by default Turbopack file system caching (beta) Optimized navigations and prefetching Improved caching APIs Build Adapters API (alpha) React [----] https://t.co/axFBZJUZpQ Next.js [--] Cache Components Turbopack enabled by default Turbopack file system caching (beta) Optimized navigations and prefetching Improved caching APIs Build Adapters API (alpha) React [----] https://t.co/axFBZJUZpQ"  
[X Link](https://x.com/wey_gu/status/1981180059567477020)  2025-10-23T02:05Z 13.4K followers, [----] engagements


"PoT Proof of Token/LLM solving puzzle blockchain 🤓 @dotey token MCP 🤣 @dotey token MCP 🤣"  
[X Link](https://x.com/wey_gu/status/1981879373600027101)  2025-10-25T00:23Z 13.4K followers, [----] engagements


"This was my first time mentoring in OSPP and I was so very lucky to had the journy with @FengzeCai Fengze was a super contributor who sole created #NebulaGraph PyG abstraction layer 👍🏻. For anyone who is hiring interns in Python Data and Web3 dev he is the one This is an article about nebula-pyg a Python library that bridges NebulaGraph and PyTorch Geometric (PyG). Today I've translated it into English and I warmly welcome everyone to use it https://t.co/CvGzk2eq8L This is an article about nebula-pyg a Python library that bridges NebulaGraph and PyTorch Geometric (PyG). Today I've"  
[X Link](https://x.com/wey_gu/status/1981924893475152317)  2025-10-25T03:24Z 13.5K followers, [----] engagements


"revisit: my recently polishing ui related things(as non-fe engineer) via codex & claude code certainly showed me the limitation of pure CLI/TUI based system. codex/claude code's advantages are: - they freshly created the swe agent from agent(not grow from copilot mode tooling) with sota model in mid-2025 in a period their agentic orchestration/runtime is better where the (context engineering) execution matters - they own the proprietary model thus could offering generous offering(forever longer) where pricing matters while there are many cases other dimension also matters in many cases that"  
[X Link](https://x.com/wey_gu/status/1983351460390940913)  2025-10-29T01:53Z 13.5K followers, [----] engagements


"i am actually surprised by composer-1 not served via cerebras/groq"  
[X Link](https://x.com/wey_gu/status/1984057113837236432)  2025-10-31T00:37Z 13.6K followers, [----] engagements


"Who is waiting for Nowledge Mem for Windows🤟 @NowledgeMem Windows"  
[X Link](https://x.com/wey_gu/status/1986323825710309700)  2025-11-06T06:44Z 13.6K followers, [----] engagements


"hallucination is a feature human memory is generative"  
[X Link](https://x.com/wey_gu/status/1987617085649015085)  2025-11-09T20:23Z 13.7K followers, [----] engagements


"Vercel anomaly detectioncodebase swe agent root cause analysis AIOps Vercels automatic anomaly detection on point. This is a glimpse of the future of the cloud. Anomaly caught AI investigation pull request / action / analysis. https://t.co/tnhg5dwOsY Vercels automatic anomaly detection on point. This is a glimpse of the future of the cloud. Anomaly caught AI investigation pull request / action / analysis. https://t.co/tnhg5dwOsY"  
[X Link](https://x.com/wey_gu/status/1989132265558012186)  2025-11-14T00:44Z 13.7K followers, [----] engagements


"@frostming90 Nowledge Memory Exchange Chrome Extension ChatGPT Atlas ChatGPT session 🤣 https://chromewebstore.google.com/detail/nowledge-memory-exchange/kjgpkgodplgakbeanoifnlpkphemcbmh https://chromewebstore.google.com/detail/nowledge-memory-exchange/kjgpkgodplgakbeanoifnlpkphemcbmh"  
[X Link](https://x.com/wey_gu/status/1991377189456212158)  2025-11-20T05:24Z 13.8K followers, [----] engagements


"@nameless_2024 Sorry thread deep chat"  
[X Link](https://x.com/wey_gu/status/1991757148473421951)  2025-11-21T06:34Z 13.8K followers, [--] engagements


"@indie_maker_fox 👍🏻"  
[X Link](https://x.com/wey_gu/status/1991800956691087420)  2025-11-21T09:28Z 13.8K followers, [---] engagements


"Linux 😭 window decoration launcher"  
[X Link](https://x.com/wey_gu/status/1992072249667694752)  2025-11-22T03:26Z 13.8K followers, [--] engagements


"I am applying similar approaches too. And not just for language model prompts but also for mid journey prompts on logo creation tasks(with images in the loop). Where its crucial for us in the loop to add essential evals/preference when feedback benchmark is not straightforward to LM. prompt prompt Midjourney prompt😉 loop preference evals My approach to writing prompts: I describe my idea to the AI and let it generate a prompt. After that I test and validate the result. Based on the validation I ask the AI to adjust the prompt then test again. I repeat this process until Im fully satisfied."  
[X Link](https://x.com/wey_gu/status/1992676960791839157)  2025-11-23T19:29Z 13.8K followers, [----] engagements


"Cursor rebased its upstream vs code"  
[X Link](https://x.com/wey_gu/status/1992783355465932902)  2025-11-24T02:32Z 13.9K followers, 10.5K engagements


"agentic idp with LlamaCloud This is a cool example of using document extraction to extract metadata for better retrieval. A big promise of using AI agents for document understanding is automatically extracting out structured data that can be indexed for much more interesting search capabilities beyond get https://t.co/4r8KiGSxwS This is a cool example of using document extraction to extract metadata for better retrieval. A big promise of using AI agents for document understanding is automatically extracting out structured data that can be indexed for much more interesting search capabilities"  
[X Link](https://x.com/wey_gu/status/1993128434542035440)  2025-11-25T01:23Z 13.8K followers, [---] engagements


"Surprisingly long time for the first time v0 added Claude Opus [---] as its model candidate the in-house model was the only options for long time and they worked great i guess its capability comes to a new level at its price. v0 Opus [---] UI pricing"  
[X Link](https://x.com/wey_gu/status/1993135516561424407)  2025-11-25T01:51Z 13.8K followers, [----] engagements


"👍🏻 Nowledge Mem Black Friday Coupon is listed @CoooolXyh @CoooolXyh"  
[X Link](https://x.com/wey_gu/status/1993234088522924402)  2025-11-25T08:23Z 13.8K followers, [----] engagements


"@levelsio is IBAN similar to ETC in China Normal cars will use ETC too in China but for taxi due to someone prefers to have a physical highway fee receipt for claim biz traffic cost thus taxi driver normally don't use ETC :D And i never saw this robot in China before today"  
[X Link](https://x.com/wey_gu/status/1993964668982055410)  2025-11-27T08:46Z 13.8K followers, [----] engagements


"My current coding stack: - ChatGPT sota + Gemini sota for deep research - deepwiki(fast and deep) for repo study - ChatWise with Kimi on Groq for ad-hoc quick search/study - Cursor agents for isolated quick study + non-ambiguous tasks like replace markdown renderer - Claude Code with Opus [---] for big feature research + dev in more chat and review summary way - Cursor Opus/Codex for iterations major trouble shooting in pair coding way editor mode - Composer for quick UI adjustments - v0 for UI design - Nowledge Mem for content exchange and management across above all Cost - Claude Max 100$ -"  
[X Link](https://x.com/wey_gu/status/1994220248422338964)  2025-11-28T01:42Z 13.9K followers, [----] engagements


"@levelsio @dwours I think its a regional food covering north vn and south cn. pho is a Chinese loan word though and in mandarin now pronounced as fen and we call it beef fen actually"  
[X Link](https://x.com/wey_gu/status/1994775685681090992)  2025-11-29T14:29Z 13.9K followers, [----] engagements


"👍 lets see when some wasm/ virtualization/ sandbox team to be acquired. Bun is joining Anthropic https://t.co/06NlXDZhNE Bun is joining Anthropic https://t.co/06NlXDZhNE"  
[X Link](https://x.com/wey_gu/status/1996005306846130306)  2025-12-02T23:55Z 13.9K followers, [----] engagements


"@levelsio If long trip scan the QR code with WeChat. To order it before onboarding had to do so on the app [-----]. This seemed to be a guide from a YouTuber. https://www.instagram.com/reel/DMgLiqxoQxg/ https://www.instagram.com/reel/DMgLiqxoQxg/"  
[X Link](https://x.com/wey_gu/status/1997598101952213058)  2025-12-07T09:24Z 14.4K followers, [----] engagements


"@metromidnightS @LotusDecoder https://ui.shadcn.com/ https://ui.shadcn.com/"  
[X Link](https://x.com/wey_gu/status/1998311758688293009)  2025-12-09T08:40Z 14.2K followers, [--] engagements


"Nowledge Mem Claude Code Plugin claude plugin marketplace add nowledge-co/community claude plugin install nowledge-mem@nowledge-community https://mem.nowledge.co/docs/integrations#claude-code Claude Code Token GitHub claude-mem Claude Code https://t.co/Wv4sDCDcGb https://mem.nowledge.co/docs/integrations#claude-code Claude Code Token GitHub claude-mem Claude Code https://t.co/Wv4sDCDcGb"  
[X Link](https://x.com/wey_gu/status/1998407360235376995)  2025-12-09T15:00Z 14.4K followers, 12.2K engagements


"Xuanwo's huge RFC impl as a SWE Agent Arena The REAL open-source Code Agent Arena. Same code context same prompts. How does your code agent work How many roundtrips does it take for your code agents to get a PR merged https://t.co/Yvy8s7tU0f https://t.co/Yu87XKh0TP The REAL open-source Code Agent Arena. Same code context same prompts. How does your code agent work How many roundtrips does it take for your code agents to get a PR merged https://t.co/Yvy8s7tU0f https://t.co/Yu87XKh0TP"  
[X Link](https://x.com/wey_gu/status/1999035902690234496)  2025-12-11T08:37Z 14.5K followers, [----] engagements


"In a long time we are building and designing on different layers due to capabilities/competence barriers among them. Those barriers assumptions are being ceased with right mindsets and toolsets. Meituan merged the frontend dev org into the backend dev org recently. Cursor team dogfooding to have everyone as peer of builders rather than dev or designers. I an infra/backend dev with cursor and v0 had been doing things I could never imagine to do in last [--] years🫡. https://twitter.com/i/web/status/1999163589350097014 https://twitter.com/i/web/status/1999163589350097014"  
[X Link](https://x.com/wey_gu/status/1999163589350097014)  2025-12-11T17:05Z 14.7K followers, [----] engagements


"macOS [----] thunderbold5 RDMA MLX RDMA over Thunderbolt New Features Enables low-latency communication between Thunderbolt [--] hosts for use cases including distributed AI inference using MLX. (164123391) https://developer.apple.com/documentation/macos-release-notes/macos-26_2-release-notes#RDMA-over-Thunderbolt https://developer.apple.com/documentation/macos-release-notes/macos-26_2-release-notes#RDMA-over-Thunderbolt"  
[X Link](https://x.com/wey_gu/status/1999650718354174023)  2025-12-13T01:20Z 14.4K followers, [----] engagements


"@Stanley_Wind proxy cheap"  
[X Link](https://x.com/wey_gu/status/1999858663192920424)  2025-12-13T15:07Z 14.4K followers, [---] engagements


"Nvidia groq game changer"  
[X Link](https://x.com/wey_gu/status/2004003727070671168)  2025-12-25T01:38Z 14.4K followers, [--] engagements


"@OnlyXuanwo @conductor_build sure indeed Installed conductor not fully leveraged though mainly due to i used massive parallel small tasks (some times from mobile phone) with cursor agent rather than Claude code"  
[X Link](https://x.com/wey_gu/status/2005597179122196812)  2025-12-29T11:10Z 14.5K followers, [---] engagements


"As we talked i don't use lagcy design tools due to i don't know how to and v0-like agentic UI building tool is my design tools. Framework's component limit how UI can be designed while for new UI designers their limitations are a gift. Or specially you should start from shadcn/UI for basic layout/flow design as the DSL between you and the LLM/Agentic UI builder(v0/glitch/lovable). Then if you are using GPUI or other framework translate to them as the target if you have to. Go or other places to see lot more UI designs(better to check different UIs but not specially on your domain as it might"  
[X Link](https://x.com/wey_gu/status/2007807633521361381)  2026-01-04T13:33Z 14.5K followers, [----] engagements


"How i design and build UI @OnlyXuanwo @yetone As we talked i don't use lagcy design tools due to i don't know how to and v0-like agentic UI building tool is my design tools. Framework's component limit how UI can be designed while for new UI designers their limitations are a gift. Or specially you should start from @OnlyXuanwo @yetone As we talked i don't use lagcy design tools due to i don't know how to and v0-like agentic UI building tool is my design tools. Framework's component limit how UI can be designed while for new UI designers their limitations are a gift. Or specially you should"  
[X Link](https://x.com/wey_gu/status/2007807744234111486)  2026-01-04T13:34Z 14.5K followers, [---] engagements


"@OnlyXuanwo @NowledgeMem 🫡🥰 UI design is really the hardest part building things"  
[X Link](https://x.com/wey_gu/status/2007810585850007758)  2026-01-04T13:45Z 14.6K followers, 11.4K engagements


"@2011sk8_ wise call btw kimi-cli's shell tool is platform aware it'll use powershell for equivalent calls while bash for macOS/Linux env"  
[X Link](https://x.com/wey_gu/status/2008131428819906960)  2026-01-05T11:00Z 14.5K followers, [---] engagements


"@willmcgugan omg pickle + pork belly + starch noodles soup is from my hometown region(north east china) that pickle is quite similar to German pickle i am surprised there is no similar soup in Germany"  
[X Link](https://x.com/wey_gu/status/2008361178066813146)  2026-01-06T02:13Z 14.5K followers, [---] engagements


"@thaddeusjiangzh e2e coverage profile pr branch"  
[X Link](https://x.com/wey_gu/status/2009159588776722784)  2026-01-08T07:05Z 14.6K followers, [---] engagements


"@OnlyXuanwo @v0 Its browser based now or still gpui"  
[X Link](https://x.com/wey_gu/status/2009238066935074964)  2026-01-08T12:17Z 14.6K followers, [---] engagements


"yeah there are many things to be done but i am so excited to witness you doing them i can see firstly we can leverage that as the standard to improve basic alignment/contract issue of the GPUI version that can help polish your basic design principles as a doc to be referred in all future UI iteration also you can even end up come with some basic "components" from them. Tasks using GPUI is really something i never imagined but i really consider it's worth the attempts and you will make it"  
[X Link](https://x.com/wey_gu/status/2009240104238502064)  2026-01-08T12:25Z 14.6K followers, [---] engagements


"Thats also why I kept a mirror/twin version of my app in v0. When I would like to iterate/thinking over building new features or flows I fork it in v0 and build and polish it there. Then if a version of UI design/new feature/refactor finalized I cherry-pick the related component to my code base or share the working thread to the FE hacker in my team. I shared this workflow in my session with TikTok too will polish its slides and compose a blog later soon https://twitter.com/i/web/status/2011236663134236715 https://twitter.com/i/web/status/2011236663134236715"  
[X Link](https://x.com/wey_gu/status/2011236663134236715)  2026-01-14T00:39Z 14.7K followers, [--] engagements


"https://t.co/D4fwmeGYKY https://t.co/D4fwmeGYKY"  
[X Link](https://x.com/wey_gu/status/2014265842687754447)  2026-01-22T09:16Z 14.7K followers, [----] engagements


"😍 Vibe DJI Mic+Spokenly+Nowledge-Mem Vibe DJI Mic+Spokenly+Nowledge-Mem"  
[X Link](https://x.com/wey_gu/status/2014957544201155068)  2026-01-24T07:04Z 14.8K followers, [----] engagements


"deepseek new model MiniMax M2.5 Qwentransformer PROpenRouter GLM-5 kimi kimi [---]  Step-3.5-flash GLM-4.7-flash homelab https://twitter.com/i/web/status/2021534570081062964 https://twitter.com/i/web/status/2021534570081062964"  
[X Link](https://x.com/anyuser/status/2021534570081062964)  2026-02-11T10:39Z 15.2K followers, 14.4K engagements


"Nowledge Mem - agent ai tool - - Nowledge Mem AI Now Mem AI Native Agent Cowork Kimi CLI - -  v0.6 AI @passluo @wey_gu @elliotchen100 nowledge @passluo @wey_gu @elliotchen100 nowledge"  
[X Link](https://x.com/anyuser/status/2022574723998122206)  2026-02-14T07:32Z 15.2K followers, [----] engagements


"twitter compaction as a service  evaluation #NebulaGraph Apache KVrocks🫡 🎉 Excited to share that our paper "CaaS-LSM: Compaction-as-a-Service for LSM-based Key-Value Stores in Storage Disaggregated Infrastructure" has been accepted at the SIGMOD [----] Research Track. 1/3 🎉 Excited to share that our paper "CaaS-LSM: Compaction-as-a-Service for LSM-based Key-Value Stores in Storage Disaggregated Infrastructure" has been accepted at the SIGMOD [----] Research Track. 1/3"  
[X Link](https://x.com/wey_gu/status/1791396129499975980)  2024-05-17T09:11Z 15.2K followers, 11K engagements


"LlamaCPP @UnslothAI cowork agentic  Qwen4 50B size  @NowledgeMem AI Now glm-4.7-flash 30B-3B A size mlx community 4bit/8bit lm studio  homelab [----] x [--] FP8 agentic https://t.co/DUnAuXB0LF glm-4.7-flash 30B-3B A size mlx community 4bit/8bit lm studio  homelab [----] x [--] FP8 agentic https://t.co/DUnAuXB0LF"  
[X Link](https://x.com/wey_gu/status/2014283724226281915)  2026-01-22T10:27Z 15.2K followers, [----] engagements


"./llama.cpp/llama-server -hf unsloth/GLM-4.7-Flash-GGUF:UD-Q4_K_XL --host 0.0.0.0 --alias "unsloth/GLM-4.7-Flash" --threads [---] --fit on --seed [----] --temp [---] --top-k [--] --top-p [---] --min-p [----] --dry-multiplier [---] --ctx-size [------] --jinja"  
[X Link](https://x.com/wey_gu/status/2015090043774529814)  2026-01-24T15:51Z 15.2K followers, [---] engagements


"the bar has been raised https://t.co/8zdy6i3cte https://t.co/LXh3YtFpxE https://t.co/8zdy6i3cte https://t.co/LXh3YtFpxE"  
[X Link](https://x.com/wey_gu/status/2017181055107973552)  2026-01-30T10:20Z 14.9K followers, [----] engagements


"Just use vokabry Prompt engineering OpenAI https://t.co/AX1Bobgsnr https://t.co/FMX7dWVDZM Prompt engineering OpenAI https://t.co/AX1Bobgsnr https://t.co/FMX7dWVDZM"  
[X Link](https://x.com/wey_gu/status/2017475272749285685)  2026-01-31T05:49Z 14.9K followers, [----] engagements


"@yihong0618 [--] coding agent"  
[X Link](https://x.com/wey_gu/status/2017478322033471753)  2026-01-31T06:01Z 14.9K followers, [----] engagements


"@XianyuLi @Cydiar404"  
[X Link](https://x.com/wey_gu/status/2017538656177954869)  2026-01-31T10:01Z 14.9K followers, [---] engagements


"qwen3-next sparse attention coder 80B A3B minimax [---] glm [---] yeah we have been working on this small coder qwen3-coder-next based on qwen3-next for effectiveness and efficiency and also preparation for the next gen models and now we opensource it for you to play with it. i hope that this can be your fast coder and best local coding yeah we have been working on this small coder qwen3-coder-next based on qwen3-next for effectiveness and efficiency and also preparation for the next gen models and now we opensource it for you to play with it. i hope that this can be your fast coder and best"  
[X Link](https://x.com/wey_gu/status/2018848724462809214)  2026-02-04T00:46Z 15.1K followers, 11.5K engagements


"@yyw2000 Wow it's connecting RAM as vRAM on Nvidia devices"  
[X Link](https://x.com/wey_gu/status/2018876239591928005)  2026-02-04T02:36Z 15K followers, [----] engagements


"3090 unsloth batch decoding [--] token/sec Agentic llama cpptemplate xml llama cpp  qwen3-next sparse attention coder 80B A3B minimax [---] glm [---]  qwen3-next sparse attention coder 80B A3B minimax [---] glm 4.7"  
[X Link](https://x.com/wey_gu/status/2018924558150258936)  2026-02-04T05:48Z 15.2K followers, [----] engagements


"@dabit3 @cognition @DevinAI @windsurf congrats"  
[X Link](https://x.com/wey_gu/status/2019186304530178362)  2026-02-04T23:08Z 15K followers, [---] engagements


"@eightnoteight yeah my go to fast-yet-working model is still k2 over groq now"  
[X Link](https://x.com/wey_gu/status/2019237943001641239)  2026-02-05T02:33Z 15K followers, [--] engagements


"@huntermbown haha seemed that we need to navigate the the homepage of the tweet to have the btn shown"  
[X Link](https://x.com/wey_gu/status/2019260931923165216)  2026-02-05T04:04Z 15K followers, [---] engagements


"openai api AI ip clash AI openai/anthropic  ClaudeMax  ClaudeMax"  
[X Link](https://x.com/wey_gu/status/2019984475678208435)  2026-02-07T04:00Z 15.1K followers, 15.2K engagements


"@ivanfioravanti for me llamacpp was with bugs/degration two days ago comparing to mlx"  
[X Link](https://x.com/wey_gu/status/2020203197231780060)  2026-02-07T18:29Z 15.2K followers, [---] engagements

Limited data mode. Full metrics available with subscription: lunarcrush.com/pricing

@wey_gu Avatar @wey_gu Wey Gu 古思为

Wey Gu 古思为 posts on X about ai, agentic, open ai, code the most. They currently have [------] followers and [---] posts still getting attention that total [---------] engagements in the last [--] hours.

Engagements: [---------] #

Engagements Line Chart

  • [--] Week [---------] +29,932%
  • [--] Month [---------] +43,446%
  • [--] Months [---------] +12,031%
  • [--] Year [---------] +13,638%

Mentions: [--] #

Mentions Line Chart

  • [--] Months [--] +119%
  • [--] Year [--] +71%

Followers: [------] #

Followers Line Chart

  • [--] Week [------] +1.10%
  • [--] Month [------] +3.50%
  • [--] Months [------] +19%
  • [--] Year [------] +30%

CreatorRank: [---------] #

CreatorRank Line Chart

Social Influence

Social category influence technology brands stocks social networks finance currencies countries travel destinations gaming

Social topic influence ai, agentic #548, open ai, code, kimi #314, claude code #1138, openclaw, 6969, llm, azure

Top accounts mentioned or mentioned by @onlyxuanwo @nowledgemem @zhangjintao9020 @thaddeusjiangzh @frostming90 @yihong0618 @yyhxiaolang @yetone @weygu @leaskh @strrlthedev @grok @danja @levelsio @passluo @elliotchen100 @danbri @dreamylostcn @jerryjliu0 @pdmproject

Top assets mentioned Alphabet Inc Class A (GOOGL) Fusion (FSN) StarLink (STARL) Cloudflare, Inc. (NET)

Top Social Posts

Top posts by engagements in the last [--] hours

"Elixir Redis server http Redis universal http 100👉 https://docs.statetrace.com/blog/redis-server/ https://docs.statetrace.com/blog/redis-server/"
X Link 2021-11-12T16:25Z 11.4K followers, [---] engagements

"While this isn't 100% true though I am looking forward to how far it could be very soon #ChatGPT knows #NebulaGraph"
X Link 2022-12-01T11:05Z 11.3K followers, [--] engagements

"#RaycastApp Pro now"
X Link 2023-05-10T08:27Z 11.4K followers, [----] engagements

"With This is possible 👍🏻 from langchain.llms import OpenLLM llm = OpenLLM.for_model(model_name='databricks/dolly-v2-3b') llm("What is the difference between a duck and a goose") http://github.com/bentoml/OpenLLM http://github.com/bentoml/OpenLLM"
X Link 2023-06-13T01:41Z 11K followers, [----] engagements

"@wjsvector @yevazn Yes and the whole infra/stack is open-source too thus you ideally could just fork it and create your own with it. 👉🏻"
X Link 2023-07-09T16:08Z [----] followers, [--] engagements

"Last week I witnessed Azure OpenAI dropped gpt-3 support(got [---] error) and one of my streamlit online demo was broken. It was interesting that previously azure [---] turbo didnt work perfectly thus I was stick on [--] for long time while in recent version its w/o issues(phew)"
X Link 2023-08-12T02:42Z [----] followers, [--] engagements

"@appinn What if there is a OpenDAL Desktop/Mobile SIG/SubProject @OnlyXuanwo"
X Link 2023-08-18T11:08Z [----] followers, [----] engagements

"@danja @danbri Its so very honored seeing OG (linked data) hacker like you trying my demo/NebulaGraph.🫡 Something could be wrong from metad which ended up Auth error was the open file limit changed from [----] please And sorry for the no hosts error too :p"
X Link 2023-08-19T23:31Z [----] followers, [--] engagements

"@danbri @danja It's OpenAI GPT Quota that ran out"
X Link 2023-08-20T11:29Z 10.2K followers, [--] engagements

". @Gephi is an amazing GraphViz tool I created #NebulaGraph #Gephi to one click to get gexf file of any query then drop the file into Gephi/Gephi-Lite for graph algo& viz If you are using docker ext its already included just update to try it out"
X Link 2023-09-02T01:22Z 10.2K followers, [----] engagements

"TIL #NebulaGraph now comes with a decent #scala and #ZIO wrapper project thanks to @dreamylostcn loved that logo"
X Link 2023-09-06T11:11Z 10.2K followers, [----] engagements

"@danja With docker or Python venv there will be no worries or burden on dependencies of system Python :)"
X Link 2023-09-15T22:15Z 10.3K followers, [--] engagements

"Excited to delve into Graph #NebulaGraph and Graph RAG with LLM at #DockerCon within the @Docker Community Will you be around in California Join me in LA; let's explore and learn together 👩💻🔍📊 Check out more details: https://bit.ly/dc23speakers https://bit.ly/dc23speakers"
X Link 2023-09-27T07:47Z 11.3K followers, 28.4K engagements

"This is a great of KG(#NebulaGraph) and LLM experiment summary and tutorial with valuable takeaways Thanks @wenqi_glantz and @jerryjliu0 Was working on things in private last few weeks hopefully I could share/upstream more work soon on this"
X Link 2023-09-30T01:34Z [----] followers, [----] engagements

"Code Interpreter API and Retrieval API again push down more work to the OpenAI/Cloud which is expected to be the model side(WHY NOT). 🤩"
X Link 2023-11-06T08:40Z [----] followers, [----] engagements

"@beihuo ASML IaaS case"
X Link 2023-11-29T01:26Z [----] followers, [----] engagements

"Millions of new materials discovered with Graph deep learning - Google DeepMind #Graph Networks for Materials Exploration (GNoME) deep learning tool dramatically increases the speed and efficiency of discovery by predicting the stability of new materials"
X Link 2023-11-30T05:55Z [----] followers, [---] engagements

"@genedna 😭"
X Link 2023-12-03T01:21Z [----] followers, [--] engagements

"It took me quite some hours to make #RedAlert2 run on my #Chromebook smoothly on a 4K screen. I put how-to together in this gist so you don't have to. [--] Chromebook crossover/Windows VM"
X Link 2023-12-31T05:14Z 10.3K followers, [----] engagements

"Happy new year my friends:)"
X Link 2024-01-01T13:58Z [--] followers, [----] engagements

"@wusheng1108 @ASFSkyWalking @zioscala ping scala hacker @dreamylostcn"
X Link 2024-01-03T13:17Z [----] followers, [---] engagements

"Python JIT in [----] https://t.co/tsWumVjZDR https://t.co/tsWumVjZDR"
X Link 2024-01-03T18:40Z 10.2K followers, [--] engagements

"Found in @ayakaneko #NebulaGraph community"
X Link 2024-01-05T01:32Z 10.3K followers, [----] engagements

"@imwsl90 @bartfastt M1 MacBook pro m1 max dock displaylink cpu dock logo"
X Link 2024-01-05T03:27Z 10.3K followers, [--] engagements

"@meterscao palm pixi"
X Link 2024-01-05T06:43Z [--] followers, [--] engagements

"🤯 When a "Modernize #NebulaGraph-Python with PDM" issue was created and the next min the creator of @pdm_project @frostming90 just magically showed(together with @yihong0618 ) And [--] min later a PR was shipped with the whole project being polished"
X Link 2024-01-05T10:18Z [--] followers, [----] engagements

"@frostming90 @pdm_project @yihong0618 Aha the time-stamp in the issue thread was twisted so I didn't realize that. Hahaha I like the saying of @yihong0618 's webhook like Spider-Sense of Spider-Man"
X Link 2024-01-05T10:40Z [--] followers, [---] engagements

"@yyhxiaolang 2080ti 22GiB RTX Titan😭"
X Link 2024-01-05T15:01Z [--] followers, [--] engagements

"@yyhxiaolang openstack cyborg Nvidia vGPU license unlock crack"
X Link 2024-01-05T18:23Z 10.2K followers, [--] engagements

"@LeaskH founding team bold YouTube clicks iPhone youtuber co-founder"
X Link 2024-01-06T02:41Z 10.2K followers, [--] engagements

"@LeaskH palm/blackberry pro and con hightlights"
X Link 2024-01-06T02:46Z 10.2K followers, [--] engagements

"@MiniCssExtract @ChristEd930427 windows mactype mactype"
X Link 2024-01-06T02:51Z [----] followers, [--] engagements

"@LeaskH walk through list cmdtab"
X Link 2024-01-06T03:01Z 10.3K followers, [--] engagements

"@NovaIntrovert"
X Link 2024-01-06T09:11Z [--] followers, [--] engagements

"@yyhxiaolang 👍"
X Link 2024-01-07T09:48Z [--] followers, [---] engagements

"@github @laixintao 🎉"
X Link 2024-01-07T16:10Z [----] followers, [--] engagements

"@sxyazi Congrats"
X Link 2024-01-08T09:06Z [----] followers, [---] engagements

"@limbopeng terminal OpenStack flake8 [--] black [--] max line 100"
X Link 2024-01-17T10:47Z 10.3K followers, [--] engagements

"@thecalicastle @yyhxiaolang :-D VueJS"
X Link 2024-01-25T06:57Z 10.2K followers, [--] engagements

"@yyhxiaolang cc @thecalicastle 🥹 useEffect VueJS 🤣"
X Link 2024-01-29T10:57Z 10.3K followers, [--] engagements

"@sxyazi congrats"
X Link 2024-02-01T10:15Z 10.3K followers, [--] engagements

"I can't wait to see Apple Vision Pro become the size of Google Glass one day"
X Link 2024-02-04T01:19Z 10.3K followers, [----] engagements

"@yihong0618 Patpat - OpenAI Azure API IP iAP apple id plus - Colab NebulaGraph-Lite block ** Colab 😭"
X Link 2024-03-06T07:38Z 10.4K followers, [--] engagements

"@zhangjintao9020 @yihong0618 Google Play Plus session 😢 plus API"
X Link 2024-03-06T10:16Z 10.4K followers, [--] engagements

"@beautyyuyanli 🥳 team #NebulaGraph GenAI will physically union next week"
X Link 2024-03-21T12:07Z 10.4K followers, [--] engagements

"scala java guobin scala hacker🫡"
X Link 2024-04-03T04:50Z 10.4K followers, [--] engagements

"@oran_ge Graph RAG LlamaIndex workshop - - - - https://bit.ly/graph-rag-workshop https://www.siwei.io/demo-dumps/local-llm/Graph_RAG_Local.html https://www.siwei.io/demo-dumps/graph-rag/GraphRAG.html https://www.siwei.io/tutors/GraphRAG/101.html https://bit.ly/graph-rag-workshop https://www.siwei.io/demo-dumps/local-llm/Graph_RAG_Local.html https://www.siwei.io/demo-dumps/graph-rag/GraphRAG.html https://www.siwei.io/tutors/GraphRAG/101.html"
X Link 2024-04-30T00:42Z 10.5K followers, [----] engagements

"Check out my X card from moimoi @moimoi_app 🎉 #moimoi #digitalmosaic"
X Link 2024-04-30T12:46Z 10.5K followers, [----] engagements

"Graph RAG with GNN. A framework for evaluating Retrieval-Augmented Text Generation for #LLMs & Graph #RAG grows in complexity It can be organized paradigm into categories SURGE uses #GNN to extract subgraphs from #knowledgegraph achieving customization #EmergingTech #AI https://t.co/Nj0Zhg3N2d https://t.co/cmy4zWpsR9 A framework for evaluating Retrieval-Augmented Text Generation for #LLMs & Graph #RAG grows in complexity It can be organized paradigm into categories SURGE uses #GNN to extract subgraphs from #knowledgegraph achieving customization #EmergingTech #AI https://t.co/Nj0Zhg3N2d"
X Link 2024-04-30T14:13Z 10.5K followers, [----] engagements

"Some of my old Graph RAG stuff is worth a retweet"
X Link 2024-04-30T14:18Z 10.5K followers, [--] engagements

"🧠 Access Distributed Graph Data(#NebulaGraph) in the great OpenDAL manner This will be sick😱. Thanks @OnlyXuanwo and Su Yan Gege for making this possible A joint collaboration & opportunity OPFS Apache OpenDAL and @NebulaGraph Hello everyone. Here's an announcement from the #OpenDAL community regarding participation in the Open Source Promotion event. This year we are featuring three projects: OPFS monoio and @NebulaGraph's support Feel free to visit https://t.co/U1i7EThdMC for more details. Hello everyone. Here's an announcement from the #OpenDAL community regarding participation in the"
X Link 2024-04-30T15:03Z 10.5K followers, [----] engagements

"@JustinLin610 Thanks"
X Link 2024-05-01T07:09Z 10.5K followers, [---] engagements

"@9hills LiteLLM Azure OAI OAI Function Calling"
X Link 2024-05-07T03:03Z 10.5K followers, [--] engagements

"Dannys work makes dev on macOS whole next level comparing to wsl2/docker desktop on windows. https://github.com/microsoft/WSL/issues/9555 In the next version of OrbStack: true near-native file system performance 210x faster than before pnpm: 0.5% overhead vs. native macOS (avg [----] vs [-----] sec) with [----] packages https://t.co/wSQIUMvbLQ https://github.com/microsoft/WSL/issues/9555 In the next version of OrbStack: true near-native file system performance 210x faster than before pnpm: 0.5% overhead vs. native macOS (avg [----] vs [-----] sec) with [----] packages https://t.co/wSQIUMvbLQ"
X Link 2024-05-13T11:04Z 10.6K followers, [----] engagements

"chatgpt-4o streaming glass"
X Link 2024-05-15T03:55Z 10.6K followers, [--] engagements

"@strrlthedev hypervisor vsphereesxi"
X Link 2024-05-21T04:58Z 10.6K followers, [--] engagements

"@strrlthedev track license ESXi vsphere"
X Link 2024-05-21T05:20Z 10.6K followers, [--] engagements

"options - raycast ad-hoc cli helper raycast ai - GitHub copilot cli - cusor terminal commad+k"
X Link 2024-05-21T05:36Z 10.6K followers, [--] engagements

"@yhcnux warpOpt + Click Tmux select Terminal"
X Link 2024-05-21T06:27Z 10.6K followers, [--] engagements

"#NebulaGraph + Apache DAL nodejs client ArkTS Palm WebOS"
X Link 2024-05-22T12:02Z 10.6K followers, [--] engagements

"@TheYotg Thanks for sharing this The GitHub page url is [---] though. https://mlx-graphs.github.io/mlx-graphs/ https://mlx-graphs.github.io/mlx-graphs/"
X Link 2024-06-06T15:05Z 10.7K followers, [---] engagements

"@faouzia You the best in this performance. As one big fan of Justin Timberlake its sooo enjoyable listening to your masterpiece version"
X Link 2024-06-14T18:07Z 10.7K followers, [---] engagements

"I was always planning to leverage the amazing @textualizeio to create a terminal #NebulaGraph UI tonight I had some time to see @tedconbeer 's great the SQL IDE now to be a Database IDE after supporting @NebulaGraph). $ pip install harlequin-nebulagraph $ harlequin -a nebulagraph demo video in the thread. https://github.com/wey-gu/harlequin-nebulagraph http://harlequin.sh http://harlequin.sh https://github.com/wey-gu/harlequin-nebulagraph http://harlequin.sh http://harlequin.sh"
X Link 2024-06-17T14:21Z 10.7K followers, [----] engagements

"Azure OpenAI(Azure OAI) #NebulaGraph team Azure OAI serving/client OAI tricky OAI function calling LiteLLM OAI function calling"
X Link 2024-06-25T05:55Z 10.7K followers, [--] engagements

"@xsser_w IP IP Reputation 😭 OpenAI traffic https call"
X Link 2024-06-25T11:09Z 10.7K followers, [--] engagements

"psw3c vue principle tailwind css v0/shadcn empower 🥹"
X Link 2024-07-07T00:52Z 10.8K followers, [--] engagements

"A great milestone. LlamaCloud is solving the key problem s elegantly for GenAI workflows. Or I should say knowledge/data driven workflow. Im excited to announce a beta release of LlamaCloud 🦙⛅ - the data processing layer for your LLM applications. Weve opened up a waitlist for users to signup To build a better knowledge assistant you need an advanced data processing layer to prevent the garbage in = garbage https://t.co/zk43Tg3CgO Im excited to announce a beta release of LlamaCloud 🦙⛅ - the data processing layer for your LLM applications. Weve opened up a waitlist for users to signup To"
X Link 2024-07-09T17:10Z 10.8K followers, [----] engagements

"@vanillaCitron 🙏"
X Link 2024-07-17T09:53Z 10.8K followers, [---] engagements

"The abstraction of LlamaIndex is data/knowledge-driven based on which we could build GenAI apps with ease and peace. The way that I use/learn LlamaIndex is to walkthrough both docs and code(of course by contributing to it) While for some who enjoyed reading in a real book there is one for you😊 and now I am thinking its a good start for rag-101 too(I used to suggest new dev to walkthrough LI docs to get started on RAG) I liked how it was structured in the book. Thanks @Vinishkakalra for sending me the preview copy."
X Link 2024-07-19T11:20Z 10.9K followers, [----] engagements

"WE ARE HIRING @beautyyuyanli #NebulaGraph GenAI - build Graph + GenAI Magic in public - scale the magic to other wizards/witches to help them do the same - hack because we can - change the world with #NebulaGraph #ISOGQL and #OpenSource Join us We are seeking hackers based in Shanghai: - Frontend hacker - Backend hacker(Python) Reach out to me or our recruiter @eva_wuwuwu please"
X Link 2024-07-19T11:54Z 10.9K followers, 27.9K engagements

"@OnlyXuanwo happy birthday xuanwo"
X Link 2024-07-21T20:48Z 10.9K followers, [---] engagements

"- copilot ++ - block selection change/apply - chat sota - sonnet gpt-4 chat apply - composer tier cursor"
X Link 2024-08-06T04:30Z 10.9K followers, [--] engagements

"@kelseyhightower highly recommend to try orbstack as replacement of both VMware fusion and docker desktop"
X Link 2024-08-19T01:31Z 11.3K followers, [----] engagements

"I have an intel PixelBook with quite good battery life apart from all my other Mac machines its perfectly designed and performs good enough for some ad-hoc dev work. The only thing I struggled with is my muscle memory of key mappings. I used to use Ubuntu desktop as daily driver for years when electron-based apps were not that popular as its today but only quite few use cases struggled already(launcher like raycast/alfred meeting app for screen sharing)"
X Link 2024-08-19T19:23Z 11.3K followers, [---] engagements

"Were thrilled to announce that @repsiace is joining the #NebulaGraph GenAI team Excited to see the amazing things well build together. Welcome aboard 🚀 🚀 Excited to start my journey at #NebulaGraph today transitioning from Rust to Python for GenAI 🚀 Excited to start my journey at #NebulaGraph today transitioning from Rust to Python for GenAI"
X Link 2024-09-10T00:33Z 11.3K followers, [----] engagements

"metas xr glass seems to be in shape I dreamed to have in google glass"
X Link 2024-09-26T12:47Z 11.3K followers, [---] engagements

"Along with Yanli and PsiACE we're so very happy to welcome to the NebulaGraph GenAI Team 😄 https://github.com/boyyangzai https://github.com/boyyangzai"
X Link 2024-10-10T06:05Z 11.2K followers, [----] engagements

"@yuyinws @nuxt_hub @stackblitz Great work Leo If you are interested in persisting pkg-graph in Graph Database to enable more graph query/ graph insight we are happy to sponsor you via #NebulaGraph Cloud"
X Link 2024-10-12T03:31Z 11.2K followers, [---] engagements

"@yuyinws @nuxt_hub @stackblitz I meant other than getting a subgraph from one node from the whole graph compute graph algorithms and persist features back to the graph and visualize based on graph algo(node importance community detection etc.)"
X Link 2024-10-12T03:35Z 11.2K followers, [---] engagements

"Now I can code (w/ cursor) even on a plane w/o starlink🤣. Qwen [---] Coder 32B on mlx_community thanks to super @Prince_Canuma Thanks 🙏 Let me try it now https://t.co/r5fINv611s Qwen [---] Coder 32B on mlx_community thanks to super @Prince_Canuma Thanks 🙏 Let me try it now https://t.co/r5fINv611s"
X Link 2024-11-12T13:37Z 11.3K followers, [----] engagements

"@zhangjintao9020 @thekonginc"
X Link 2024-11-15T10:01Z 11.6K followers, [---] engagements

"@Akinsanmidev @buaaxhm drop your question here or in a github issue/stackoverflow post"
X Link 2024-11-21T02:30Z 11.4K followers, [--] engagements

"So very exciting for Qwen2-VL support and MLX. Coming this month: a complete rewrite of the Ollama engine. Includes CUDA support and more models. Coming in 2025: MLX support https://t.co/PcdA1xtxvQ Coming this month: a complete rewrite of the Ollama engine. Includes CUDA support and more models. Coming in 2025: MLX support https://t.co/PcdA1xtxvQ"
X Link 2024-12-03T03:34Z 11.4K followers, [---] engagements

"GitHub [----] 😋"
X Link 2024-12-17T12:21Z 11.6K followers, [----] engagements

"2024 Raycast wrapped"
X Link 2024-12-18T02:06Z 11.6K followers, [---] engagements

"o1 preview was slow yet truly valuable. o1 is already fast deepseek r1 and qwen qwq are open-weight Gemini thinking is reasoning better yet visible. Now we have o3 and o3-mini in new level of reasoning. Reasoning models to be norm as agentic backbone is soon to come🥳 Day 12: Early evals for OpenAI o3 (yes we skipped a number) https://t.co/iWXg9IGuZM Day 12: Early evals for OpenAI o3 (yes we skipped a number) https://t.co/iWXg9IGuZM"
X Link 2024-12-20T23:15Z 11.6K followers, [---] engagements

"@OikawaRizumu @vuejsnation"
X Link 2024-12-24T10:41Z 11.6K followers, [---] engagements

"Wow thanks qwen team merry Christmas 🎄 Happy holidays This is something fun for you to play with. QVQ the new model can think and reason on images 🎄 Happy holidays This is something fun for you to play with. QVQ the new model can think and reason on images"
X Link 2024-12-24T17:26Z 11.6K followers, [---] engagements

"notebookLM 🤯 I created an open-source implementation of @notebooklm 🤖 Deepseek-V3 API using @OpenRouterAI 🎙 @play_ht TTS using @FAL API 📚 Create AI podcasts on ANY topic 100% Customizable All this in [--] lines of code Check out the GitHub repo: https://t.co/Jp96O5aBci https://t.co/ivqk8rYNaS 🤯 I created an open-source implementation of @notebooklm 🤖 Deepseek-V3 API using @OpenRouterAI 🎙 @play_ht TTS using @FAL API 📚 Create AI podcasts on ANY topic 100% Customizable All this in [--] lines of code Check out the GitHub repo: https://t.co/Jp96O5aBci https://t.co/ivqk8rYNaS"
X Link 2025-01-04T09:54Z 11.6K followers, [---] engagements

"The GenAI-based content creation engine refly is source-available today a great call on this:) https://github.com/refly-ai/refly https://github.com/refly-ai/refly"
X Link 2025-01-24T10:07Z 11.7K followers, [---] engagements

"@jerryjliu0 @llama_index @databricks @KPMG congrats"
X Link 2025-05-01T22:41Z 12K followers, [--] engagements

"Interesting lexical patterns analysis based clustering showing the distilling lineages between latest models. R1-0528 showed closer to Gemini while R1 to OpenAI. DeepSeek-R1-0528 Gemini R1 OpenAI DeepSeek R1 Google Gemini R1 OpenAI llama https://t.co/Qr1L5c5Jid DeepSeek-R1-0528 Gemini R1 OpenAI DeepSeek R1 Google Gemini R1 OpenAI llama https://t.co/Qr1L5c5Jid"
X Link 2025-05-31T01:32Z 12.1K followers, [---] engagements

"source code is the new target in GenAI era. design docs meta source code and their change histories are true sources now AI Coding AI Coding LLM AI Coding AI Coding AI Coding LLM AI Coding"
X Link 2025-06-05T16:32Z 12.2K followers, [---] engagements

"Cua introduced windows sandbox support🫡 computer = Computer( provider_type=VMProviderType.WINSANDBOX os_type="windows" memory="4GB" ) Introducing Windows Sandbox support - run computer-use agents on Windows business apps without VMs or cloud costs. 1/6 https://t.co/uL6LOMifAs Introducing Windows Sandbox support - run computer-use agents on Windows business apps without VMs or cloud costs. 1/6 https://t.co/uL6LOMifAs"
X Link 2025-06-19T17:38Z 12.3K followers, [---] engagements

"@frostming90 @dinah_zhang @Piglei"
X Link 2025-06-25T02:45Z 12.3K followers, [--] engagements

"This work showed RL transfers SFT barely does. RL generalizes SFT doesn't. Self-exploration and feedback leads to AGI not SFT. Does that mean once we reach certain intelligence level (able to generate valid candidate) the value of human generated data is diminishing. And feedback especially rule generated massive RL generalizes SFT doesn't. Self-exploration and feedback leads to AGI not SFT. Does that mean once we reach certain intelligence level (able to generate valid candidate) the value of human generated data is diminishing. And feedback especially rule generated massive"
X Link 2025-07-03T02:36Z 12.3K followers, [---] engagements

"@lantianlaoli openai-translator"
X Link 2025-07-05T19:06Z 12.4K followers, [---] engagements

"@zhangjintao9020 @himself_65 @strrlthedev dxt client eval claude desktop/ cursor/ chatwise client shared sandbox nodejs runtime"
X Link 2025-07-15T10:50Z 12.4K followers, [---] engagements

"chatwise + groq k2 + websearch(llama on groq) ad-hoc search k2 on groq would be a killing existence for certain usage k2 on groq would be a killing existence for certain usage"
X Link 2025-07-15T11:03Z 12.4K followers, [----] engagements

"Still more works to do #buildinpublic 😭 #buildinpublic https://t.co/sBdSkIS0FV 😭 #buildinpublic https://t.co/sBdSkIS0FV"
X Link 2025-07-18T02:15Z 12.5K followers, [----] engagements

"@zhangjintao9020 Sort of But not to touch the Note-App area just help sink to/explore from our key insights threads across Precious TALKS with GenAI agents"
X Link 2025-07-18T02:35Z 12.4K followers, [---] engagements

"Hey @grok Who are my potential ones to follow from my 2-hop top mutuals but not yet connected yet with my most shared top mutuals"
X Link 2025-07-19T02:04Z 12.5K followers, [---] engagements

"@grok @yetone @OnlyXuanwo @oran_ge @YannLecun @lexfridman @ylecun @drfeifei Yeah so I know you are lacking of api expressions to actually do such query right"
X Link 2025-07-19T02:11Z 12.5K followers, [--] engagements

"@grok @yetone @OnlyXuanwo @YannLecun @lexfridman @ylecun @drfeifei @karpathy Thanks for the transparency"
X Link 2025-07-19T02:25Z 12.5K followers, [--] engagements

"@grok @OikawaRizumu @QuantumTransf @Nickel_Angel @shiratsuyudachi @Meowneko_Mio @JakkuSakura @Eureka222323 @kiligulyanov @RobertoC_cr03x @misaka__network @saonanaa @kz152cx @sCuSO_4 @ggccsxhn @Exusiai159 @isaac_eins @aqCuSO_4 @InfectQWQ @ChinTAT_2008 @vanto156607 @AFeiSigniFy__ @bghtnya @yqua_ @bushi_8964 @sxshuishuimao @qingshixiang258 @hiner233 @Chiyuyu520 @Miisago_han @xiaoyieQAQ @VicoOci @mitsuki87613 @yunliu2005 @All12240519 @jiangkedou @ilkos_sdvx @Nickel_Schwarz @dimangmu8964 @89JinPing64 @DDzhanshou5150 @FoxRed695449 @Cyan_0731 @lcMenci @ProbiusOfficial @mike_cheng1208 @Stv_Lynn"
X Link 2025-07-19T04:54Z 12.5K followers, [----] engagements

"for me it's still cursor. my workflow is more pair coding with cursor and pair/vibe design with v0 dot dev. in study phase i talked to gemini pro/o4 mini/claude over my upstream projects within github copilot; run deep research over chatgpt/gemini; then i started writing RFC within the repo back and forth the study and poc code with cursor's agent to finalize feasibility and arch design etc. then i'll have main arch/feature design and i ask cursor to sync things during my iteration of design. and sinking/updating cursor rules from those design docs(keeps sharp and short). then i started to"
X Link 2025-07-19T09:10Z 12.9K followers, 15.4K engagements

"For me its more pair coding rather than vibe coding now when building things. @yetone for me it's still cursor. my workflow is more pair coding with cursor and pair/vibe design with v0 dot dev. in study phase i talked to gemini pro/o4 mini/claude over my upstream projects within github copilot; run deep research over chatgpt/gemini; then i started writing RFC @yetone for me it's still cursor. my workflow is more pair coding with cursor and pair/vibe design with v0 dot dev. in study phase i talked to gemini pro/o4 mini/claude over my upstream projects within github copilot; run deep research"
X Link 2025-07-19T12:39Z 12.5K followers, [---] engagements

"@suavesu Omg didnt know the license switch from gplv3 to apache2"
X Link 2025-07-20T06:39Z 12.5K followers, [---] engagements

"Now I am running @nowledgelabs at night and my free time only.(alpha launch soon) Hopefully will make it my daytime soon Still more works to do #buildinpublic https://t.co/dgkB5KXWCF Still more works to do #buildinpublic https://t.co/dgkB5KXWCF"
X Link 2025-07-22T10:48Z 12.5K followers, [----] engagements

"original precision [--] gib ram 🫡 You can now run gpt-oss-120b & 20b locally with our GGUFs 🦥 Run OpenAI's 120b model on 66GB RAM & 20b model on 14GB RAM. Both in original precision. Uploads includes our chat template fixes. Guide: https://t.co/LZPNUvyy2G GGUF: https://t.co/mfZDpekRER https://t.co/VUmkjjs5ZX You can now run gpt-oss-120b & 20b locally with our GGUFs 🦥 Run OpenAI's 120b model on 66GB RAM & 20b model on 14GB RAM. Both in original precision. Uploads includes our chat template fixes. Guide: https://t.co/LZPNUvyy2G GGUF: https://t.co/mfZDpekRER https://t.co/VUmkjjs5ZX"
X Link 2025-08-06T01:29Z 12.7K followers, [---] engagements

"We need an universal well-known paths to cache those LLM/Embedding models on-device"
X Link 2025-08-13T07:36Z 12.7K followers, [----] engagements

"@frostming90 Ollama is heavy yet not support mlx/onnx yet for now"
X Link 2025-08-13T08:58Z 12.7K followers, [---] engagements

"@OnlyXuanwo @tison1096 +1 for dingtalk and huawei im"
X Link 2025-08-13T13:19Z 12.7K followers, [---] engagements

"@hylarucoder gpt5 troubleshooting"
X Link 2025-08-15T07:39Z 12.7K followers, [--] engagements

"@repsiace we are chain of wechat messages"
X Link 2025-08-15T08:39Z 12.7K followers, [--] engagements

"@localhost_4173 ping @JustinLin610"
X Link 2025-08-20T10:25Z 12.8K followers, [---] engagements

"@ZinkLu2 @nowledgelabs remote llm 🫡🧎"
X Link 2025-08-21T04:15Z 12.8K followers, [---] engagements

"we could define the term vibe here further to me i do pair coding rather than vibe coding while pair coding with genai is a just the form of broader definition of vide coding in most context . vibe coding . vibe coding"
X Link 2025-08-21T11:01Z 12.8K followers, [---] engagements

"https://github.com/deepseek-ai/DeepSeek-V3/issues/849 https://github.com/deepseek-ai/DeepSeek-V3/issues/849"
X Link 2025-08-26T23:37Z 12.9K followers, [--] engagements

"semtools with claude code Turn Claude Code into a Financial Analyst 🤖💹 In this video we point Claude Code at a bucket of 10k filing PDFs and have it perform complex analysis across the entire set of docs Claude Code doesnt have file understanding out of the box (it kind of does but its terrible / https://t.co/N5EuyJw5PS Turn Claude Code into a Financial Analyst 🤖💹 In this video we point Claude Code at a bucket of 10k filing PDFs and have it perform complex analysis across the entire set of docs Claude Code doesnt have file understanding out of the box (it kind of does but its terrible /"
X Link 2025-09-01T20:32Z 12.9K followers, [----] engagements

"@hd_nvim This was in our mind too while just didnt have chance to actually impl. Define proper graph schema code parsing mechanism and finally agentic operators to leverage the graph should help and ideally it could scales yet latency wise to be performant"
X Link 2025-09-06T06:55Z 13K followers, [---] engagements

"SGLang mooncake integration out of box support We are excited to announce SGLang HiCache our community solution for hierarchical KV caching to power high-performance LLM serving. ⚡ Performance: up to [--] throughput and 80% TTFT reduction demonstrated in benchmarks and real-world deployments. 🗂 Flexibility: seamless https://t.co/9pPjFbuGuT We are excited to announce SGLang HiCache our community solution for hierarchical KV caching to power high-performance LLM serving. ⚡ Performance: up to [--] throughput and 80% TTFT reduction demonstrated in benchmarks and real-world deployments. 🗂 Flexibility:"
X Link 2025-09-12T10:02Z 13K followers, [----] engagements

"@shinyzhu cloudflare"
X Link 2025-09-20T02:42Z 13K followers, [---] engagements

"@Kechen_Li2018 nuxtjs remix svelte I am not with fe background actually"
X Link 2025-09-24T14:36Z 13.1K followers, [--] engagements

"TIL Brahmand an opensource stateless graph query/computation layer on ClickHouse that speaks OpenCypher. https://github.com/darshanDevrai/brahmand Introducing Brahmand: an open source Graph Engine built on top of ClickHouse Extending ClickHouse with graph modeling and OpenCypher merging OLAP speed with graph analysis. Brahmand is a high-performance stateless graph-analysis layer for ClickHouse. Brahmand is targeting https://t.co/JS3RQzgXsB https://github.com/darshanDevrai/brahmand Introducing Brahmand: an open source Graph Engine built on top of ClickHouse Extending ClickHouse with graph"
X Link 2025-09-25T01:05Z 13.1K followers, [----] engagements

"Kwais coding model looks amazingly good in its size. 🚀Introducing KAT-Dev-32B & KAT-Coder -KAT-Dev-32B62.4% on SWE-Bench Verified top-5 among open-source models -KAT-Coder73.4% on SWE-Bench Verified rivals top proprietary models 👇 Try it 🤗 Model: https://t.co/IF20Ptd8FF 🔗 Blog: https://t.co/021ZrW7caQ https://t.co/U8mYXkqpiV 🚀Introducing KAT-Dev-32B & KAT-Coder -KAT-Dev-32B62.4% on SWE-Bench Verified top-5 among open-source models -KAT-Coder73.4% on SWE-Bench Verified rivals top proprietary models 👇 Try it 🤗 Model: https://t.co/IF20Ptd8FF 🔗 Blog: https://t.co/021ZrW7caQ"
X Link 2025-09-29T00:40Z 13.1K followers, [----] engagements

"Apache Gravitino makes datalake wirable to agentic stacks 👍 🚀 Apache Gravitino 1.0.0 is here Metadata Goes Agentic. From Metadata Management to Contextual Engineering: ⚡ Stats Policy Job systems 🤖 MCP server (agent-ready) 🔒 RBAC governance 🔗 https://t.co/DyOcGHB4tT #OpenSource #AI #Metadata #Datastrato #ApacheGravitino https://t.co/6wUsgekiVe 🚀 Apache Gravitino 1.0.0 is here Metadata Goes Agentic. From Metadata Management to Contextual Engineering: ⚡ Stats Policy Job systems 🤖 MCP server (agent-ready) 🔒 RBAC governance 🔗 https://t.co/DyOcGHB4tT #OpenSource #AI #Metadata #Datastrato"
X Link 2025-10-01T23:47Z 13.1K followers, [---] engagements

"Thanks for the shoutout @chatmcp Thrilled to integrate Nowledge Mem (impressed on how seamlessly it was done and the great UX of into the MCP ecosystemempowering local agents with graph-augmented memory To celebrate use promo code MCPSO95 for an almost-free upgrade to Nowledge Mem Pro (limited time) Demo: Join alpha waitlist: Intro blog: Let's build smarter knowledge flywheels together #AI #KnowledgeGraph #MCP https://nowled.ge/blog-intro-mem https://nowled.ge/alpha https://x.com/nowledgelabs/status/1952578790775017977 http://mcp.so http://mcp.so Great to have Nowledge Mem MCP on board"
X Link 2025-10-02T16:08Z 13.2K followers, 13K engagements

"@hd_nvim @OpenAI"
X Link 2025-10-07T09:34Z 13.2K followers, [---] engagements

"Forgetting is a gift. ChatGPT can now automatically manage your saved memoriesno more memory full. You can also search and sort memories by recency and choose which to re-prioritize in settings. Rolling out to Plus and Pro users on the web globally starting today. https://t.co/T1vSNH5289 https://t.co/xRHLFTu2Am ChatGPT can now automatically manage your saved memoriesno more memory full. You can also search and sort memories by recency and choose which to re-prioritize in settings. Rolling out to Plus and Pro users on the web globally starting today. https://t.co/T1vSNH5289"
X Link 2025-10-16T01:51Z 13.3K followers, [---] engagements

"a k8s native approach is on the way. Both OpenEdx and tutor community are quite active and friendly👍 I will for sure explore more And I am happy to have a chance to contribute to the community:"
X Link 2021-05-01T03:16Z 13.8K followers, [--] engagements

"1. Fixed some invalid commands inthe openedx native deployment documentation [--]. Fixed LMS uwsgi [---] error when request header size exceeds 4096K PR: https://github.com/overhangio/tutor/pull/427 https://openedx.atlassian.net/wiki/spaces/OpenOPS/pages/60227913/Managing+Open+edX+Tips+and+Tricks https://github.com/overhangio/tutor/pull/427 https://openedx.atlassian.net/wiki/spaces/OpenOPS/pages/60227913/Managing+Open+edX+Tips+and+Tricks"
X Link 2021-05-01T03:20Z 13.8K followers, [--] engagements

"@strrlthedev opps just go to model providers to know this too 🤣"
X Link 2025-05-30T03:49Z 14.7K followers, [---] engagements

"Cursor is one of the few implements prompts rather than only tools of the MCP protocol. Thus in cursor you can call @NowledgeMem in the chat to summarize then memorize your key insights in the thread. And all we need is to type / then mem itll focus on prompt from Nowledge Mem MCP and enter to call it. Now cursor allows us to map this flow to ANY prompt in user space. make your own commands /fix /refactor /plan soon in @cursor_ai ☺ make your own commands /fix /refactor /plan soon in @cursor_ai ☺"
X Link 2025-09-05T16:38Z 13.8K followers, [----] engagements

". @NowledgeMem was built to bring us the freedom of Agents/AI Tools build our own context flywheel😉 OpenAI Agent SDK OpenAI API Demo IP OpenAI ChatGPT OpenAI Agent SDK OpenAI API Demo IP OpenAI ChatGPT"
X Link 2025-10-16T07:09Z 13.3K followers, [----] engagements

"Upgraded from NextJS [--] to 16.0.0: - local dev build is way better(not that good as vite though) - production build is way better(4min30sec to 1min28sec) Next.js [--] Cache Components Turbopack enabled by default Turbopack file system caching (beta) Optimized navigations and prefetching Improved caching APIs Build Adapters API (alpha) React [----] https://t.co/axFBZJUZpQ Next.js [--] Cache Components Turbopack enabled by default Turbopack file system caching (beta) Optimized navigations and prefetching Improved caching APIs Build Adapters API (alpha) React [----] https://t.co/axFBZJUZpQ"
X Link 2025-10-23T02:05Z 13.4K followers, [----] engagements

"PoT Proof of Token/LLM solving puzzle blockchain 🤓 @dotey token MCP 🤣 @dotey token MCP 🤣"
X Link 2025-10-25T00:23Z 13.4K followers, [----] engagements

"This was my first time mentoring in OSPP and I was so very lucky to had the journy with @FengzeCai Fengze was a super contributor who sole created #NebulaGraph PyG abstraction layer 👍🏻. For anyone who is hiring interns in Python Data and Web3 dev he is the one This is an article about nebula-pyg a Python library that bridges NebulaGraph and PyTorch Geometric (PyG). Today I've translated it into English and I warmly welcome everyone to use it https://t.co/CvGzk2eq8L This is an article about nebula-pyg a Python library that bridges NebulaGraph and PyTorch Geometric (PyG). Today I've"
X Link 2025-10-25T03:24Z 13.5K followers, [----] engagements

"revisit: my recently polishing ui related things(as non-fe engineer) via codex & claude code certainly showed me the limitation of pure CLI/TUI based system. codex/claude code's advantages are: - they freshly created the swe agent from agent(not grow from copilot mode tooling) with sota model in mid-2025 in a period their agentic orchestration/runtime is better where the (context engineering) execution matters - they own the proprietary model thus could offering generous offering(forever longer) where pricing matters while there are many cases other dimension also matters in many cases that"
X Link 2025-10-29T01:53Z 13.5K followers, [----] engagements

"i am actually surprised by composer-1 not served via cerebras/groq"
X Link 2025-10-31T00:37Z 13.6K followers, [----] engagements

"Who is waiting for Nowledge Mem for Windows🤟 @NowledgeMem Windows"
X Link 2025-11-06T06:44Z 13.6K followers, [----] engagements

"hallucination is a feature human memory is generative"
X Link 2025-11-09T20:23Z 13.7K followers, [----] engagements

"Vercel anomaly detectioncodebase swe agent root cause analysis AIOps Vercels automatic anomaly detection on point. This is a glimpse of the future of the cloud. Anomaly caught AI investigation pull request / action / analysis. https://t.co/tnhg5dwOsY Vercels automatic anomaly detection on point. This is a glimpse of the future of the cloud. Anomaly caught AI investigation pull request / action / analysis. https://t.co/tnhg5dwOsY"
X Link 2025-11-14T00:44Z 13.7K followers, [----] engagements

"@frostming90 Nowledge Memory Exchange Chrome Extension ChatGPT Atlas ChatGPT session 🤣 https://chromewebstore.google.com/detail/nowledge-memory-exchange/kjgpkgodplgakbeanoifnlpkphemcbmh https://chromewebstore.google.com/detail/nowledge-memory-exchange/kjgpkgodplgakbeanoifnlpkphemcbmh"
X Link 2025-11-20T05:24Z 13.8K followers, [----] engagements

"@nameless_2024 Sorry thread deep chat"
X Link 2025-11-21T06:34Z 13.8K followers, [--] engagements

"@indie_maker_fox 👍🏻"
X Link 2025-11-21T09:28Z 13.8K followers, [---] engagements

"Linux 😭 window decoration launcher"
X Link 2025-11-22T03:26Z 13.8K followers, [--] engagements

"I am applying similar approaches too. And not just for language model prompts but also for mid journey prompts on logo creation tasks(with images in the loop). Where its crucial for us in the loop to add essential evals/preference when feedback benchmark is not straightforward to LM. prompt prompt Midjourney prompt😉 loop preference evals My approach to writing prompts: I describe my idea to the AI and let it generate a prompt. After that I test and validate the result. Based on the validation I ask the AI to adjust the prompt then test again. I repeat this process until Im fully satisfied."
X Link 2025-11-23T19:29Z 13.8K followers, [----] engagements

"Cursor rebased its upstream vs code"
X Link 2025-11-24T02:32Z 13.9K followers, 10.5K engagements

"agentic idp with LlamaCloud This is a cool example of using document extraction to extract metadata for better retrieval. A big promise of using AI agents for document understanding is automatically extracting out structured data that can be indexed for much more interesting search capabilities beyond get https://t.co/4r8KiGSxwS This is a cool example of using document extraction to extract metadata for better retrieval. A big promise of using AI agents for document understanding is automatically extracting out structured data that can be indexed for much more interesting search capabilities"
X Link 2025-11-25T01:23Z 13.8K followers, [---] engagements

"Surprisingly long time for the first time v0 added Claude Opus [---] as its model candidate the in-house model was the only options for long time and they worked great i guess its capability comes to a new level at its price. v0 Opus [---] UI pricing"
X Link 2025-11-25T01:51Z 13.8K followers, [----] engagements

"👍🏻 Nowledge Mem Black Friday Coupon is listed @CoooolXyh @CoooolXyh"
X Link 2025-11-25T08:23Z 13.8K followers, [----] engagements

"@levelsio is IBAN similar to ETC in China Normal cars will use ETC too in China but for taxi due to someone prefers to have a physical highway fee receipt for claim biz traffic cost thus taxi driver normally don't use ETC :D And i never saw this robot in China before today"
X Link 2025-11-27T08:46Z 13.8K followers, [----] engagements

"My current coding stack: - ChatGPT sota + Gemini sota for deep research - deepwiki(fast and deep) for repo study - ChatWise with Kimi on Groq for ad-hoc quick search/study - Cursor agents for isolated quick study + non-ambiguous tasks like replace markdown renderer - Claude Code with Opus [---] for big feature research + dev in more chat and review summary way - Cursor Opus/Codex for iterations major trouble shooting in pair coding way editor mode - Composer for quick UI adjustments - v0 for UI design - Nowledge Mem for content exchange and management across above all Cost - Claude Max 100$ -"
X Link 2025-11-28T01:42Z 13.9K followers, [----] engagements

"@levelsio @dwours I think its a regional food covering north vn and south cn. pho is a Chinese loan word though and in mandarin now pronounced as fen and we call it beef fen actually"
X Link 2025-11-29T14:29Z 13.9K followers, [----] engagements

"👍 lets see when some wasm/ virtualization/ sandbox team to be acquired. Bun is joining Anthropic https://t.co/06NlXDZhNE Bun is joining Anthropic https://t.co/06NlXDZhNE"
X Link 2025-12-02T23:55Z 13.9K followers, [----] engagements

"@levelsio If long trip scan the QR code with WeChat. To order it before onboarding had to do so on the app [-----]. This seemed to be a guide from a YouTuber. https://www.instagram.com/reel/DMgLiqxoQxg/ https://www.instagram.com/reel/DMgLiqxoQxg/"
X Link 2025-12-07T09:24Z 14.4K followers, [----] engagements

"@metromidnightS @LotusDecoder https://ui.shadcn.com/ https://ui.shadcn.com/"
X Link 2025-12-09T08:40Z 14.2K followers, [--] engagements

"Nowledge Mem Claude Code Plugin claude plugin marketplace add nowledge-co/community claude plugin install nowledge-mem@nowledge-community https://mem.nowledge.co/docs/integrations#claude-code Claude Code Token GitHub claude-mem Claude Code https://t.co/Wv4sDCDcGb https://mem.nowledge.co/docs/integrations#claude-code Claude Code Token GitHub claude-mem Claude Code https://t.co/Wv4sDCDcGb"
X Link 2025-12-09T15:00Z 14.4K followers, 12.2K engagements

"Xuanwo's huge RFC impl as a SWE Agent Arena The REAL open-source Code Agent Arena. Same code context same prompts. How does your code agent work How many roundtrips does it take for your code agents to get a PR merged https://t.co/Yvy8s7tU0f https://t.co/Yu87XKh0TP The REAL open-source Code Agent Arena. Same code context same prompts. How does your code agent work How many roundtrips does it take for your code agents to get a PR merged https://t.co/Yvy8s7tU0f https://t.co/Yu87XKh0TP"
X Link 2025-12-11T08:37Z 14.5K followers, [----] engagements

"In a long time we are building and designing on different layers due to capabilities/competence barriers among them. Those barriers assumptions are being ceased with right mindsets and toolsets. Meituan merged the frontend dev org into the backend dev org recently. Cursor team dogfooding to have everyone as peer of builders rather than dev or designers. I an infra/backend dev with cursor and v0 had been doing things I could never imagine to do in last [--] years🫡. https://twitter.com/i/web/status/1999163589350097014 https://twitter.com/i/web/status/1999163589350097014"
X Link 2025-12-11T17:05Z 14.7K followers, [----] engagements

"macOS [----] thunderbold5 RDMA MLX RDMA over Thunderbolt New Features Enables low-latency communication between Thunderbolt [--] hosts for use cases including distributed AI inference using MLX. (164123391) https://developer.apple.com/documentation/macos-release-notes/macos-26_2-release-notes#RDMA-over-Thunderbolt https://developer.apple.com/documentation/macos-release-notes/macos-26_2-release-notes#RDMA-over-Thunderbolt"
X Link 2025-12-13T01:20Z 14.4K followers, [----] engagements

"@Stanley_Wind proxy cheap"
X Link 2025-12-13T15:07Z 14.4K followers, [---] engagements

"Nvidia groq game changer"
X Link 2025-12-25T01:38Z 14.4K followers, [--] engagements

"@OnlyXuanwo @conductor_build sure indeed Installed conductor not fully leveraged though mainly due to i used massive parallel small tasks (some times from mobile phone) with cursor agent rather than Claude code"
X Link 2025-12-29T11:10Z 14.5K followers, [---] engagements

"As we talked i don't use lagcy design tools due to i don't know how to and v0-like agentic UI building tool is my design tools. Framework's component limit how UI can be designed while for new UI designers their limitations are a gift. Or specially you should start from shadcn/UI for basic layout/flow design as the DSL between you and the LLM/Agentic UI builder(v0/glitch/lovable). Then if you are using GPUI or other framework translate to them as the target if you have to. Go or other places to see lot more UI designs(better to check different UIs but not specially on your domain as it might"
X Link 2026-01-04T13:33Z 14.5K followers, [----] engagements

"How i design and build UI @OnlyXuanwo @yetone As we talked i don't use lagcy design tools due to i don't know how to and v0-like agentic UI building tool is my design tools. Framework's component limit how UI can be designed while for new UI designers their limitations are a gift. Or specially you should start from @OnlyXuanwo @yetone As we talked i don't use lagcy design tools due to i don't know how to and v0-like agentic UI building tool is my design tools. Framework's component limit how UI can be designed while for new UI designers their limitations are a gift. Or specially you should"
X Link 2026-01-04T13:34Z 14.5K followers, [---] engagements

"@OnlyXuanwo @NowledgeMem 🫡🥰 UI design is really the hardest part building things"
X Link 2026-01-04T13:45Z 14.6K followers, 11.4K engagements

"@2011sk8_ wise call btw kimi-cli's shell tool is platform aware it'll use powershell for equivalent calls while bash for macOS/Linux env"
X Link 2026-01-05T11:00Z 14.5K followers, [---] engagements

"@willmcgugan omg pickle + pork belly + starch noodles soup is from my hometown region(north east china) that pickle is quite similar to German pickle i am surprised there is no similar soup in Germany"
X Link 2026-01-06T02:13Z 14.5K followers, [---] engagements

"@thaddeusjiangzh e2e coverage profile pr branch"
X Link 2026-01-08T07:05Z 14.6K followers, [---] engagements

"@OnlyXuanwo @v0 Its browser based now or still gpui"
X Link 2026-01-08T12:17Z 14.6K followers, [---] engagements

"yeah there are many things to be done but i am so excited to witness you doing them i can see firstly we can leverage that as the standard to improve basic alignment/contract issue of the GPUI version that can help polish your basic design principles as a doc to be referred in all future UI iteration also you can even end up come with some basic "components" from them. Tasks using GPUI is really something i never imagined but i really consider it's worth the attempts and you will make it"
X Link 2026-01-08T12:25Z 14.6K followers, [---] engagements

"Thats also why I kept a mirror/twin version of my app in v0. When I would like to iterate/thinking over building new features or flows I fork it in v0 and build and polish it there. Then if a version of UI design/new feature/refactor finalized I cherry-pick the related component to my code base or share the working thread to the FE hacker in my team. I shared this workflow in my session with TikTok too will polish its slides and compose a blog later soon https://twitter.com/i/web/status/2011236663134236715 https://twitter.com/i/web/status/2011236663134236715"
X Link 2026-01-14T00:39Z 14.7K followers, [--] engagements

"https://t.co/D4fwmeGYKY https://t.co/D4fwmeGYKY"
X Link 2026-01-22T09:16Z 14.7K followers, [----] engagements

"😍 Vibe DJI Mic+Spokenly+Nowledge-Mem Vibe DJI Mic+Spokenly+Nowledge-Mem"
X Link 2026-01-24T07:04Z 14.8K followers, [----] engagements

"deepseek new model MiniMax M2.5 Qwentransformer PROpenRouter GLM-5 kimi kimi [---] Step-3.5-flash GLM-4.7-flash homelab https://twitter.com/i/web/status/2021534570081062964 https://twitter.com/i/web/status/2021534570081062964"
X Link 2026-02-11T10:39Z 15.2K followers, 14.4K engagements

"Nowledge Mem - agent ai tool - - Nowledge Mem AI Now Mem AI Native Agent Cowork Kimi CLI - - v0.6 AI @passluo @wey_gu @elliotchen100 nowledge @passluo @wey_gu @elliotchen100 nowledge"
X Link 2026-02-14T07:32Z 15.2K followers, [----] engagements

"twitter compaction as a service evaluation #NebulaGraph Apache KVrocks🫡 🎉 Excited to share that our paper "CaaS-LSM: Compaction-as-a-Service for LSM-based Key-Value Stores in Storage Disaggregated Infrastructure" has been accepted at the SIGMOD [----] Research Track. 1/3 🎉 Excited to share that our paper "CaaS-LSM: Compaction-as-a-Service for LSM-based Key-Value Stores in Storage Disaggregated Infrastructure" has been accepted at the SIGMOD [----] Research Track. 1/3"
X Link 2024-05-17T09:11Z 15.2K followers, 11K engagements

"LlamaCPP @UnslothAI cowork agentic Qwen4 50B size @NowledgeMem AI Now glm-4.7-flash 30B-3B A size mlx community 4bit/8bit lm studio homelab [----] x [--] FP8 agentic https://t.co/DUnAuXB0LF glm-4.7-flash 30B-3B A size mlx community 4bit/8bit lm studio homelab [----] x [--] FP8 agentic https://t.co/DUnAuXB0LF"
X Link 2026-01-22T10:27Z 15.2K followers, [----] engagements

"./llama.cpp/llama-server -hf unsloth/GLM-4.7-Flash-GGUF:UD-Q4_K_XL --host 0.0.0.0 --alias "unsloth/GLM-4.7-Flash" --threads [---] --fit on --seed [----] --temp [---] --top-k [--] --top-p [---] --min-p [----] --dry-multiplier [---] --ctx-size [------] --jinja"
X Link 2026-01-24T15:51Z 15.2K followers, [---] engagements

"the bar has been raised https://t.co/8zdy6i3cte https://t.co/LXh3YtFpxE https://t.co/8zdy6i3cte https://t.co/LXh3YtFpxE"
X Link 2026-01-30T10:20Z 14.9K followers, [----] engagements

"Just use vokabry Prompt engineering OpenAI https://t.co/AX1Bobgsnr https://t.co/FMX7dWVDZM Prompt engineering OpenAI https://t.co/AX1Bobgsnr https://t.co/FMX7dWVDZM"
X Link 2026-01-31T05:49Z 14.9K followers, [----] engagements

"@yihong0618 [--] coding agent"
X Link 2026-01-31T06:01Z 14.9K followers, [----] engagements

"@XianyuLi @Cydiar404"
X Link 2026-01-31T10:01Z 14.9K followers, [---] engagements

"qwen3-next sparse attention coder 80B A3B minimax [---] glm [---] yeah we have been working on this small coder qwen3-coder-next based on qwen3-next for effectiveness and efficiency and also preparation for the next gen models and now we opensource it for you to play with it. i hope that this can be your fast coder and best local coding yeah we have been working on this small coder qwen3-coder-next based on qwen3-next for effectiveness and efficiency and also preparation for the next gen models and now we opensource it for you to play with it. i hope that this can be your fast coder and best"
X Link 2026-02-04T00:46Z 15.1K followers, 11.5K engagements

"@yyw2000 Wow it's connecting RAM as vRAM on Nvidia devices"
X Link 2026-02-04T02:36Z 15K followers, [----] engagements

"3090 unsloth batch decoding [--] token/sec Agentic llama cpptemplate xml llama cpp qwen3-next sparse attention coder 80B A3B minimax [---] glm [---] qwen3-next sparse attention coder 80B A3B minimax [---] glm 4.7"
X Link 2026-02-04T05:48Z 15.2K followers, [----] engagements

"@dabit3 @cognition @DevinAI @windsurf congrats"
X Link 2026-02-04T23:08Z 15K followers, [---] engagements

"@eightnoteight yeah my go to fast-yet-working model is still k2 over groq now"
X Link 2026-02-05T02:33Z 15K followers, [--] engagements

"@huntermbown haha seemed that we need to navigate the the homepage of the tweet to have the btn shown"
X Link 2026-02-05T04:04Z 15K followers, [---] engagements

"openai api AI ip clash AI openai/anthropic ClaudeMax ClaudeMax"
X Link 2026-02-07T04:00Z 15.1K followers, 15.2K engagements

"@ivanfioravanti for me llamacpp was with bugs/degration two days ago comparing to mlx"
X Link 2026-02-07T18:29Z 15.2K followers, [---] engagements

Limited data mode. Full metrics available with subscription: lunarcrush.com/pricing

@wey_gu
/creator/twitter::wey_gu