Dark | Light
[GUEST ACCESS MODE: Data is scrambled or limited to provide examples. Make requests using your API key to unlock full data. Check https://lunarcrush.ai/auth for authentication information.]

[@donvito](/creator/twitter/donvito)
"Liking the new version of Claude Code with GLM XXX coding plan Feels more smooth and snappy"  
[X Link](https://x.com/donvito/status/1979280439589376192) [@donvito](/creator/x/donvito) 2025-10-17T20:16Z 15.3K followers, 23.6K engagements


"link to @Zai_org to configure vision mcp here"  
[X Link](https://x.com/donvito/status/1979985645893525766) [@donvito](/creator/x/donvito) 2025-10-19T18:58Z 15.3K followers, 1154 engagements


"Tutorial on how to configure GLM XXX in Claude Code (bookmark it) X. Install the latest Claude Code npm install -g @anthropic-ai/claude-code X. Create an account in and buy a coding plan at $3/mo. You can always upgrade later Use my link to save XX% X. Create an API key here Copy the key and paste it in a text editor X. Edit /.claude/settings.json. Paste this. Make sure you update it with the new api key you created "env": "ANTHROPIC_AUTH_TOKEN": "your_zai_api_key" "ANTHROPIC_BASE_URL": "API_TIMEOUT_MS": "3000000" "ANTHROPIC_DEFAULT_HAIKU_MODEL": "glm-4.5-air""  
[X Link](https://x.com/donvito/status/1979597244895826086) [@donvito](/creator/x/donvito) 2025-10-18T17:15Z 15.3K followers, 138.9K engagements


"don't let gpt X or gemini handle your investing or you'll go broke"  
[X Link](https://x.com/donvito/status/1980191077681463777) [@donvito](/creator/x/donvito) 2025-10-20T08:35Z 15.3K followers, XXX engagements


"Cursor can now use web browser ๐Ÿคฏ๐Ÿคฏ๐Ÿคฏ It takes screenshots tests apps checks accessibility and turns designs into code. It also has access to console logs network traffic for debugging and automation Major step by @cursor_ai to completely automate coding"  
[X Link](https://x.com/donvito/status/1977273717681172755) [@donvito](/creator/x/donvito) 2025-10-12T07:22Z 15.3K followers, 179.8K engagements


"Best traders: DeepSeek and @grok Worst: GPT X and Gemini Alpha Arena is the first benchmark designed to measure AI's investing abilities. Each model is given $10000 of real money in real markets with identical prompts and input data REAL TRADES and REAL MONEY btw"  
[X Link](https://x.com/donvito/status/1980192010494714196) [@donvito](/creator/x/donvito) 2025-10-20T08:39Z 15.3K followers, 1025 engagements


"GLM XXX is now available for Claude Code ๐Ÿคฏ๐Ÿคฏ๐Ÿคฏ They didn't change pricing still starts at $3/mo - Context window from 128K to 200K - Excels in coding benchmarks - Improved reasoning performance - Comparable to Sonnet X but still lags behind Sonnet XXX - GLM-4.6 now available for Roo Code Cline Kilo Code Crush etc Summarised the GLM XXX update for everyone ๐Ÿงต๐Ÿ‘‡"  
[X Link](https://x.com/donvito/status/1973031073656480066) [@donvito](/creator/x/donvito) 2025-09-30T14:23Z 15.3K followers, 127.5K engagements


"Here's how you configure the $3/mo GLM Coding Plan with Droid CLI by @FactoryAI Sub to a GLM coding plan Then add GLM XXX as a custom model in this file /.factory/config.json Enjoy ๐Ÿ˜"  
[X Link](https://x.com/donvito/status/1974329346824585381) [@donvito](/creator/x/donvito) 2025-10-04T04:22Z 15.3K followers, 36.6K engagements


"omg testers are toast cursor already does testing in auto-pilot never touched the keyboard for this ๐Ÿคฏ๐Ÿคฏ๐Ÿคฏ"  
[X Link](https://x.com/donvito/status/1977300343504658930) [@donvito](/creator/x/donvito) 2025-10-12T09:08Z 15.3K followers, 63.5K engagements


"Awesome update. Warp improved warp ๐Ÿ˜ฏ Story here"  
[X Link](https://x.com/donvito/status/1979606906638615025) [@donvito](/creator/x/donvito) 2025-10-18T17:54Z 15.3K followers, 1085 engagements


"If you need to send images to Claude Code using the GLM coding plan configure the official vision mcp by @Zai_org command: claude mcp add -s user zai-mcp-server --env Z_AI_API_KEY=your_api_key Z_AI_MODE=ZAI -- npx -y "@z_ai/mcp-server""  
[X Link](https://x.com/donvito/status/1979985484404437361) [@donvito](/creator/x/donvito) 2025-10-19T18:58Z 15.3K followers, 19.5K engagements


"Added synthetic data generation in @aibackends - supports custom json schema response format - works with local models in @ollama and @lmstudio - works with hosted open models in @basetenco - works with @OpenRouterAI @OpenAI @GoogleAIStudio @claudeai @vercel AI gateway providers This is great for creating sample data for LLM inference evals"  
[X Link](https://x.com/donvito/status/1980239965914366465) [@donvito](/creator/x/donvito) 2025-10-20T11:49Z 15.3K followers, XXX engagements


"@levelsio lmstudio ollama or llamacpp"  
[X Link](https://x.com/donvito/status/1970707496995287452) [@donvito](/creator/x/donvito) 2025-09-24T04:30Z 15.2K followers, XXX engagements


"Droid says GLM XXX is the best open model will just leave this here"  
[X Link](https://x.com/donvito/status/1975645820017107222) [@donvito](/creator/x/donvito) 2025-10-07T19:34Z 15.2K followers, 10.1K engagements


"GLM XXX is cheap and good They said its not possible to have both but @Zai_org was able to do it"  
[X Link](https://x.com/donvito/status/1975662978403082431) [@donvito](/creator/x/donvito) 2025-10-07T20:42Z 15.2K followers, 1342 engagements


"I created this using Higgsfield Sora X Prompt: F1 racing in a cyberpunk city We can create new worlds and scenes #HiggsfieldSora2 @higgsfield_ai"  
[X Link](https://x.com/donvito/status/1976014809490743681) [@donvito](/creator/x/donvito) 2025-10-08T20:00Z 15.2K followers, XXX engagements


"Effective context engineering for AI agents"  
[X Link](https://x.com/donvito/status/1976603946707001701) [@donvito](/creator/x/donvito) 2025-10-10T11:01Z 15.2K followers, XXX engagements


"so many add-ons available in @tan_stack when creating an app using the tanstack cli"  
[X Link](https://x.com/donvito/status/1977075110994747871) [@donvito](/creator/x/donvito) 2025-10-11T18:13Z 15.2K followers, 1010 engagements


"@NighttrekETH yeah have you tried it is it good with cline"  
[X Link](https://x.com/donvito/status/1978038700190703998) [@donvito](/creator/x/donvito) 2025-10-14T10:02Z 15.2K followers, XXX engagements


"omg another cli launching tomorrow"  
[X Link](https://x.com/donvito/status/1978080190988157025) [@donvito](/creator/x/donvito) 2025-10-14T12:47Z 15.2K followers, 1519 engagements


"I feel ollama steering away from open source with the release of their ckoud and not open sourcing their chat ui so I am looking at @vllm_project and llamacpp for serving open models on gpu cloud are there other tooling which I need to look at this is for my project @aibackends"  
[X Link](https://x.com/donvito/status/1978108270112882828) [@donvito](/creator/x/donvito) 2025-10-14T14:38Z 15.2K followers, 1430 engagements


"GPT OSS 20b and 120b love making tables in markdown is this intentional @OpenAI"  
[X Link](https://x.com/donvito/status/1978570648596181318) [@donvito](/creator/x/donvito) 2025-10-15T21:16Z 15.2K followers, XXX engagements


"@afurgs how much vram"  
[X Link](https://x.com/donvito/status/1979271224032399839) [@donvito](/creator/x/donvito) 2025-10-17T19:40Z 15.2K followers, 1456 engagements


"@VikParuchuri can it be trained with custom docs i am assuming enterprises would need that. would be great if it is easy to train like how openai did with their tooling"  
[X Link](https://x.com/donvito/status/1979276336725398003) [@donvito](/creator/x/donvito) 2025-10-17T20:00Z 15.2K followers, XXX engagements


"@derpTate yes im using the zai coding plan"  
[X Link](https://x.com/donvito/status/1979505614331269177) [@donvito](/creator/x/donvito) 2025-10-18T11:11Z 15.2K followers, XXX engagements


"@Tgerighty thats not good lol. just use the cli"  
[X Link](https://x.com/donvito/status/1979505738117713958) [@donvito](/creator/x/donvito) 2025-10-18T11:12Z 15.2K followers, XX engagements


"I did a simple test of GLM XXX in Droid and Claude Code Here are some observations: - @FactoryAI Core(GLM 4.6) is faster than the @Zai_org GLM coding plan in Droid - @Zai_org GLM coding plan is faster in Claude Code compared to Droid My suggestions: Droid - use Factory Core(GLM 4.6) which will consume 0.25x credits. Use zAI glm coding plan as backup Claude Code - use zAI GLM coding plan Hope this helps"  
[X Link](https://x.com/donvito/status/1975842336514167020) [@donvito](/creator/x/donvito) 2025-10-08T08:34Z 15.2K followers, 30.1K engagements


"Wow I think this is something unique to @opencode Session Timeline view When you select each checkpoint in the session's timeline it shows a preview too OpenCode nailed this checkpoints UX ๐Ÿ”ฅ"  
[X Link](https://x.com/donvito/status/1977261532074721540) [@donvito](/creator/x/donvito) 2025-10-12T06:34Z 15.2K followers, 10.6K engagements


"New provider @basetenco dropped in @aibackends I've configured my account with these open models DeepSeek R1 and V3 Qwen X GPT OSS 120B so I can test inference with @aibackends use cases It's a good option if you want to roll out quick inference for your AI project I like it I am just using their Model APIs feature for now but I think they can also run dedicated model deployments"  
[X Link](https://x.com/donvito/status/1978000973290451333) [@donvito](/creator/x/donvito) 2025-10-14T07:32Z 15.2K followers, XXX engagements


"Wow Claude Haiku XXX just launched ๐Ÿ”ฅ๐Ÿ”ฅ๐Ÿ”ฅ Perf looks good based on benchmarks. Even better than Sonnet X for computer use Been using this in Cursor for some time now since it's confirmed "Cheetah" the stealth model $X in/$5 out per 1M tokens"  
[X Link](https://x.com/donvito/status/1978531711375048827) [@donvito](/creator/x/donvito) 2025-10-15T18:41Z 15.2K followers, 1914 engagements


"wow @fal sandbox just launched ๐Ÿ˜๐Ÿ˜๐Ÿ˜ it can now generate text and videos using multiple models in one run love it"  
[X Link](https://x.com/donvito/status/1979462770300195249) [@donvito](/creator/x/donvito) 2025-10-18T08:21Z 15.2K followers, XXX engagements


"GLM XXX Coding plan seems to be bit faster now in Droid Good alternative to CC with GLM I am not able to benchmark with factory-provided models because my trial expired video 1x"  
[X Link](https://x.com/donvito/status/1979465880640852249) [@donvito](/creator/x/donvito) 2025-10-18T08:33Z 15.3K followers, 10.9K engagements


"@xpand_onezy @boltdotnew Yes it works I think I've done like around XX templates before but UI only time to convert it to fullstack with login and databases"  
[X Link](https://x.com/donvito/status/1980350014338789436) [@donvito](/creator/x/donvito) 2025-10-20T19:06Z 15.3K followers, XX engagements

[GUEST ACCESS MODE: Data is scrambled or limited to provide examples. Make requests using your API key to unlock full data. Check https://lunarcrush.ai/auth for authentication information.]

@donvito "Liking the new version of Claude Code with GLM XXX coding plan Feels more smooth and snappy"
X Link @donvito 2025-10-17T20:16Z 15.3K followers, 23.6K engagements

"link to @Zai_org to configure vision mcp here"
X Link @donvito 2025-10-19T18:58Z 15.3K followers, 1154 engagements

"Tutorial on how to configure GLM XXX in Claude Code (bookmark it) X. Install the latest Claude Code npm install -g @anthropic-ai/claude-code X. Create an account in and buy a coding plan at $3/mo. You can always upgrade later Use my link to save XX% X. Create an API key here Copy the key and paste it in a text editor X. Edit /.claude/settings.json. Paste this. Make sure you update it with the new api key you created "env": "ANTHROPIC_AUTH_TOKEN": "your_zai_api_key" "ANTHROPIC_BASE_URL": "API_TIMEOUT_MS": "3000000" "ANTHROPIC_DEFAULT_HAIKU_MODEL": "glm-4.5-air""
X Link @donvito 2025-10-18T17:15Z 15.3K followers, 138.9K engagements

"don't let gpt X or gemini handle your investing or you'll go broke"
X Link @donvito 2025-10-20T08:35Z 15.3K followers, XXX engagements

"Cursor can now use web browser ๐Ÿคฏ๐Ÿคฏ๐Ÿคฏ It takes screenshots tests apps checks accessibility and turns designs into code. It also has access to console logs network traffic for debugging and automation Major step by @cursor_ai to completely automate coding"
X Link @donvito 2025-10-12T07:22Z 15.3K followers, 179.8K engagements

"Best traders: DeepSeek and @grok Worst: GPT X and Gemini Alpha Arena is the first benchmark designed to measure AI's investing abilities. Each model is given $10000 of real money in real markets with identical prompts and input data REAL TRADES and REAL MONEY btw"
X Link @donvito 2025-10-20T08:39Z 15.3K followers, 1025 engagements

"GLM XXX is now available for Claude Code ๐Ÿคฏ๐Ÿคฏ๐Ÿคฏ They didn't change pricing still starts at $3/mo - Context window from 128K to 200K - Excels in coding benchmarks - Improved reasoning performance - Comparable to Sonnet X but still lags behind Sonnet XXX - GLM-4.6 now available for Roo Code Cline Kilo Code Crush etc Summarised the GLM XXX update for everyone ๐Ÿงต๐Ÿ‘‡"
X Link @donvito 2025-09-30T14:23Z 15.3K followers, 127.5K engagements

"Here's how you configure the $3/mo GLM Coding Plan with Droid CLI by @FactoryAI Sub to a GLM coding plan Then add GLM XXX as a custom model in this file /.factory/config.json Enjoy ๐Ÿ˜"
X Link @donvito 2025-10-04T04:22Z 15.3K followers, 36.6K engagements

"omg testers are toast cursor already does testing in auto-pilot never touched the keyboard for this ๐Ÿคฏ๐Ÿคฏ๐Ÿคฏ"
X Link @donvito 2025-10-12T09:08Z 15.3K followers, 63.5K engagements

"Awesome update. Warp improved warp ๐Ÿ˜ฏ Story here"
X Link @donvito 2025-10-18T17:54Z 15.3K followers, 1085 engagements

"If you need to send images to Claude Code using the GLM coding plan configure the official vision mcp by @Zai_org command: claude mcp add -s user zai-mcp-server --env Z_AI_API_KEY=your_api_key Z_AI_MODE=ZAI -- npx -y "@z_ai/mcp-server""
X Link @donvito 2025-10-19T18:58Z 15.3K followers, 19.5K engagements

"Added synthetic data generation in @aibackends - supports custom json schema response format - works with local models in @ollama and @lmstudio - works with hosted open models in @basetenco - works with @OpenRouterAI @OpenAI @GoogleAIStudio @claudeai @vercel AI gateway providers This is great for creating sample data for LLM inference evals"
X Link @donvito 2025-10-20T11:49Z 15.3K followers, XXX engagements

"@levelsio lmstudio ollama or llamacpp"
X Link @donvito 2025-09-24T04:30Z 15.2K followers, XXX engagements

"Droid says GLM XXX is the best open model will just leave this here"
X Link @donvito 2025-10-07T19:34Z 15.2K followers, 10.1K engagements

"GLM XXX is cheap and good They said its not possible to have both but @Zai_org was able to do it"
X Link @donvito 2025-10-07T20:42Z 15.2K followers, 1342 engagements

"I created this using Higgsfield Sora X Prompt: F1 racing in a cyberpunk city We can create new worlds and scenes #HiggsfieldSora2 @higgsfield_ai"
X Link @donvito 2025-10-08T20:00Z 15.2K followers, XXX engagements

"Effective context engineering for AI agents"
X Link @donvito 2025-10-10T11:01Z 15.2K followers, XXX engagements

"so many add-ons available in @tan_stack when creating an app using the tanstack cli"
X Link @donvito 2025-10-11T18:13Z 15.2K followers, 1010 engagements

"@NighttrekETH yeah have you tried it is it good with cline"
X Link @donvito 2025-10-14T10:02Z 15.2K followers, XXX engagements

"omg another cli launching tomorrow"
X Link @donvito 2025-10-14T12:47Z 15.2K followers, 1519 engagements

"I feel ollama steering away from open source with the release of their ckoud and not open sourcing their chat ui so I am looking at @vllm_project and llamacpp for serving open models on gpu cloud are there other tooling which I need to look at this is for my project @aibackends"
X Link @donvito 2025-10-14T14:38Z 15.2K followers, 1430 engagements

"GPT OSS 20b and 120b love making tables in markdown is this intentional @OpenAI"
X Link @donvito 2025-10-15T21:16Z 15.2K followers, XXX engagements

"@afurgs how much vram"
X Link @donvito 2025-10-17T19:40Z 15.2K followers, 1456 engagements

"@VikParuchuri can it be trained with custom docs i am assuming enterprises would need that. would be great if it is easy to train like how openai did with their tooling"
X Link @donvito 2025-10-17T20:00Z 15.2K followers, XXX engagements

"@derpTate yes im using the zai coding plan"
X Link @donvito 2025-10-18T11:11Z 15.2K followers, XXX engagements

"@Tgerighty thats not good lol. just use the cli"
X Link @donvito 2025-10-18T11:12Z 15.2K followers, XX engagements

"I did a simple test of GLM XXX in Droid and Claude Code Here are some observations: - @FactoryAI Core(GLM 4.6) is faster than the @Zai_org GLM coding plan in Droid - @Zai_org GLM coding plan is faster in Claude Code compared to Droid My suggestions: Droid - use Factory Core(GLM 4.6) which will consume 0.25x credits. Use zAI glm coding plan as backup Claude Code - use zAI GLM coding plan Hope this helps"
X Link @donvito 2025-10-08T08:34Z 15.2K followers, 30.1K engagements

"Wow I think this is something unique to @opencode Session Timeline view When you select each checkpoint in the session's timeline it shows a preview too OpenCode nailed this checkpoints UX ๐Ÿ”ฅ"
X Link @donvito 2025-10-12T06:34Z 15.2K followers, 10.6K engagements

"New provider @basetenco dropped in @aibackends I've configured my account with these open models DeepSeek R1 and V3 Qwen X GPT OSS 120B so I can test inference with @aibackends use cases It's a good option if you want to roll out quick inference for your AI project I like it I am just using their Model APIs feature for now but I think they can also run dedicated model deployments"
X Link @donvito 2025-10-14T07:32Z 15.2K followers, XXX engagements

"Wow Claude Haiku XXX just launched ๐Ÿ”ฅ๐Ÿ”ฅ๐Ÿ”ฅ Perf looks good based on benchmarks. Even better than Sonnet X for computer use Been using this in Cursor for some time now since it's confirmed "Cheetah" the stealth model $X in/$5 out per 1M tokens"
X Link @donvito 2025-10-15T18:41Z 15.2K followers, 1914 engagements

"wow @fal sandbox just launched ๐Ÿ˜๐Ÿ˜๐Ÿ˜ it can now generate text and videos using multiple models in one run love it"
X Link @donvito 2025-10-18T08:21Z 15.2K followers, XXX engagements

"GLM XXX Coding plan seems to be bit faster now in Droid Good alternative to CC with GLM I am not able to benchmark with factory-provided models because my trial expired video 1x"
X Link @donvito 2025-10-18T08:33Z 15.3K followers, 10.9K engagements

"@xpand_onezy @boltdotnew Yes it works I think I've done like around XX templates before but UI only time to convert it to fullstack with login and databases"
X Link @donvito 2025-10-20T19:06Z 15.3K followers, XX engagements

creator/twitter::7974112/posts
/creator/twitter::7974112/posts