@Ali_TongyiLab Tongyi Lab

Tongyi Lab posts on X about we are, ai, build, model the most. They currently have [------] followers and [---] posts still getting attention that total [-------] engagements in the last [--] hours.

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

• [--] Week [---------] +107%
• [--] Month [---------] -25%
• [--] Months [----------] +2,682%

Mentions: [--] #

• [--] Months [---] +359%

Followers: [------] #

• [--] Week [------] +1.30%
• [--] Month [------] +10%
• [--] Months [------] +503%

CreatorRank: [------] #

Social Influence

Social category influence products 3% gaming 3% technology brands 3% social networks 3% travel destinations 1% finance 1%

Social topic influence we are #133, ai 10%, build 8%, model #359, to the 7%, creators #2030, in the 6%, check 6%, art 4%, community 4%

Top accounts mentioned or mentioned by @dx8152 @furaidosu @alitongyilab @modelscope2022 @ollama @ostrisai @yeq6x @comfyui @1aminvesting @newaiworld_ @barrakali @creator_kachun @eugeniofierro3 @renderartist @rubynian2 @aleenaamiir @machinedelusion @prithivmlmodss @0x00krypt @furkangozukara

Top Social Posts

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

"1/ [--] We are pleased to introduce Z-Image an efficient 6-billion-parameter foundation model for image generation. Through systematic optimization it proves that top-tier performance is achievable without relying on enormous model sizes delivering strong results in photorealistic generation and bilingual text rendering that are comparable to leading commercial models"
X Link 2025-11-27T06:33Z 18.5K followers, 7.8M engagements

"We put the "Turbo" in the name but you guys put us on the charts. Z-Image Turbo is sitting at the top of HF Models AND Spaces trending lists. Were not crying youre crying. (Okay were definitely crying🥹). Huge thanks to the community"
X Link 2025-11-28T02:39Z 18.6K followers, 1.5M engagements

"🤩Big shoutout to @ostrisai for the AI Toolkit integration. Support for training LoRAs for @Ali_TongyiLab Z-Image has been added to AI Toolkit. No code changes were necessary but I did add a template to the UI so you can quickly get started. https://t.co/w8EzrTDcvO Support for training LoRAs for @Ali_TongyiLab Z-Image has been added to AI Toolkit. No code changes were necessary but I did add a template to the UI so you can quickly get started. https://t.co/w8EzrTDcvO"
X Link 2026-01-28T03:09Z 18.5K followers, [----] engagements

"TheQwen-Image Global Challengeis about to officially open. Get your datasets ready. We are looking for the best stylization LoRAs in the world. High-performance rewards for high-performance creativity. Stay sharp. Details drop in a few days"
X Link 2026-01-31T04:33Z 18.4K followers, [----] engagements

"1/6 We are excited to introduce Z-Imagethe foundation model of the ⚡- Image family engineered for good quality robust generative diversity broad stylistic coverage and precise prompt adherence. While Z-Image-Turbo is built for speed Z-Image is a full-capacity undistilled transformer designed to be the backbone for creators researchers and developers who require the highest level of creative freedom. https://twitter.com/i/web/status/2016186674531758285 https://twitter.com/i/web/status/2016186674531758285"
X Link 2026-01-27T16:28Z 18.6K followers, 656.5K engagements

"1/3 Qwen3-ASR speech recognition series is now officially open-source 🚀Leveraging the innovativeAuT Speech Encoderand the powerful multimodal capabilities of theQwen3-Omnibase model we are introducing three flagship models designed to redefine accuracy and efficiency in speech processing. https://twitter.com/i/web/status/2016879001868902790 https://twitter.com/i/web/status/2016879001868902790"
X Link 2026-01-29T14:20Z 18.6K followers, 2.6M engagements

"2/3 ✨Qwen3-ASR-1.7B: The Performance Powerhouse Multilingual Dominance:Supports [--] languages/dialects; outperforms existing open-source models across [--] mainstream languages. Dialects & Accents:Covers [--] Chinese dialects and outperforms major commercial APIs. English accent recognition across [--] countries exceeds the performance of GPT-4o and Gemini. Robust in Complex Scenes:From singing voice transcription to low SNR (noise) environments and elderly/children's speechstable output guaranteed. ⚡Qwen3-ASR-0.6B: The Perfect Balance of Speed & Skill Ultra-Fast Inference:Achieves 2000x throughput"
X Link 2026-01-29T14:20Z 18.6K followers, [----] engagements

"Thank you @dx8152 for building and sharing this Do you ever rest This is another incredibly practical tool. A new LoRA for removing shadows and fixing exposure built on Qwen-Image. Download link and tutorial are in comments"
X Link 2025-11-10T02:23Z 18.6K followers, 11.2M engagements

"Love seeing what the community builds with @ModelScope2022 . @dx8152 just dropped a game-changing Light Migration LoRA for Qwen-Image-Edit-2509. It solves the "secondary lighting" headache perfectly. Incredible work. 👏"
X Link 2025-12-08T02:45Z 18.6K followers, 3.1M engagements

"Found a hardcore LoRA workflow Generate a PLY point cloud -- Adjust angle in editor --Refine with Qwen-Image-Edit-2511 Gaussian Splash LoRA. It accurately restores complex perspective shifts. As shown in the demo it handles 3D rotation and can even restore high-def details from blurry close-ups. Within a [--] range the consistency is unmatched. Huge thanks to @dx8152 for this contribution 🫡 https://twitter.com/i/web/status/2011259509516456302 https://twitter.com/i/web/status/2011259509516456302"
X Link 2026-01-14T02:10Z 18.6K followers, 1.3M engagements

"1/6 We are thrilled to announce theQwen-Image LoRA Training Competition Were calling on global developers to push the boundaries of fine-tuning using the Qwen-Image base model. Submission Window:Feb [--] March [--] [----] The challenge starts now"
X Link 2026-02-02T11:00Z 18.6K followers, 3.5M engagements

"5/6 The Rules of the Game.⚖"
X Link 2026-02-02T11:00Z 18.6K followers, [----] engagements

"6/6Don't miss out on the discussion. Join the server now 👇 https://discord.com/invite/z9QeEveYB2 https://discord.com/invite/z9QeEveYB2"
X Link 2026-02-02T11:00Z 18.6K followers, [----] engagements

"1/4IntroducingQwen3-Coder-Next: Our latest open-weights model designed specifically to power the next generation of autonomousCoding Agents. Built on Qwen3-Next this model is engineered to handle complex long-horizon programming tasks with unprecedented efficiency. High-performance agentic intelligence is now in your hands. https://twitter.com/i/web/status/2018832449992790247 https://twitter.com/i/web/status/2018832449992790247"
X Link 2026-02-03T23:42Z 18.6K followers, 152.7K engagements

"3/4🚀 Try it now: HuggingFace: ModelScope: https://modelscope.cn/collections/Qwen/Qwen3-Coder-Next https://huggingface.co/collections/Qwen/qwen3-coder-next https://modelscope.cn/collections/Qwen/Qwen3-Coder-Next https://huggingface.co/collections/Qwen/qwen3-coder-next"
X Link 2026-02-03T23:42Z 18.6K followers, [----] engagements

"4/4Don't miss out on the discussion. Join the server now https://discord.com/invite/z9QeEveYB2 https://discord.com/invite/z9QeEveYB2"
X Link 2026-02-03T23:42Z 18.6K followers, [----] engagements

"Big shoutout to developer for this incredible deep dive into the Z-Image hybrid workflowThis tutorial perfectly demonstrates how to master the balance between speed and hyper-realism. By blending Z-Image Turbo for composition and Base for intricate skin textures you get the best of both worlds: efficiency and pore-level detail.Check out the full breakdown below and start creating 📺 Watch here: https://www.youtube.com/watchv=FOshPE-Qr_4 https://www.youtube.com/watchv=FOshPE-Qr_4"
X Link 2026-02-04T06:17Z 18.6K followers, 10.3K engagements

"Big thanks to our featured creator @dx8152 for dropping Gauss Splash LoRAa slick Track [--] entry that reimagines image perspective control 🎨 Their behind-the-scenes video shows the full training process on @ModelScope2022 from idea to LoRA. A great guide for anyone getting started Want to build your own Join the Qwen-Image LoRA Hackathon free compute real prizes (iPhone PS5) and open until March [--] 👉 #HappyQwenday #QwenImageLora https://modelscope.ai/active/qwenimagelora LoRA Training Contest: Win iPhone/PS5 and other prizes Gaussian splash LoRA creation process shared. Competition"
X Link 2026-02-06T03:13Z 18.6K followers, [----] engagements

"Hello developers This week were thrilled to launch the Qwen-Image LoRA Training Competition a global call to fine-tune innovate and build with purpose. And while you train weve got new tools to supercharge your workflow. Lets dive in. https://open.substack.com/pub/tongyilab/p/feb-6-2026-the-tongyi-weeklyr=6te3ta&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true https://open.substack.com/pub/tongyilab/p/feb-6-2026-the-tongyi-weeklyr=6te3ta&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true"
X Link 2026-02-06T08:01Z 18.6K followers, [----] engagements

"Huge thanks to @Render_Artist for sharing the Track [--] Featured Creator project: a LoRA that transforms photos into black-and-white sketch-style coloring booksperfect for creativity mindfulness and accessibility Inspired You can join too https://modelscope.ai/active/qwenimagelora Thrilled to be a Featured Creator for the Qwen-Image LoRA Online Hackathon Ive partnered with @Ali_TongyiLab and @ModelScope2022 to share my process for training a LoRA for the category of "AI for Good" (Track 2). You can train your own Qwen LoRA 100% for free on ModelScope https://t.co/D4cqy0YuM5"
X Link 2026-02-09T07:56Z 18.6K followers, [----] engagements

"Bravo to @Rubynian2 for her brilliant Medieval Marginalia LoRAtransforming ordinary images into whimsical manuscript-style art straight out of the Middle Ages 🏰 ✨As a Featured Creator shes shown how LoRAs can turn historical inspiration into powerful creative toolsand her step-by-step video makes it accessible to all. Join the competition and build your own: https://modelscope.ai/active/qwenimagelora I am beyond excited to join the Qwen-Image LoRA Hackathon as a Featured Creator to share my latest obsession: Medieval Marginalia Competition Information: https://t.co/LERGYTav9E Medieval_Petify"
X Link 2026-02-09T09:13Z 18.6K followers, [----] engagements

"Zvec is here. The SQLite of Vector Databases. 🚀 An open-sourceEmbedded Vector Databasebuilt for the edge. Lightweight: Zvec is a zero-dependency embedded vector database that runs locally with minimal resource usage ideal for edge devices. Lightning-fast: Over [----] QPSmore than [--] the previous leaderboard #1 (ZillizCloud)while also substantially reducing index build time. RAG-Ready: Zvec natively supports dynamic CRUD hybrid search multi-vector fusion and rerankingenabling robust on-device RAG. Try it now: 🌐Website: 📝Blog: https://zvec.org/en/blog/introduction/ https://zvec.org/en/"
X Link 2026-02-09T14:00Z 18.6K followers, 18K engagements

"A huge thank you to Bijan Bowen for this thorough deep dive into Qwen3-Coder-Next 🚀 Want to see our new open-weights model in action and understand its impact on the coding agent landscape If you are looking to learn more about its real-world performance and technical breakthroughs this video is a must-watch. 📺 Watch the full review here: https://www.youtube.com/watchv=uxvttLUOY-k https://www.youtube.com/watchv=uxvttLUOY-k"
X Link 2026-02-10T04:13Z 18.6K followers, [----] engagements

"1/4 🚀We are launching Qwen-Image-2.0 a next-generation foundational image generation model. The key highlights of Qwen-Image-2.0 include: Professional Typography Rendering: Supports 1k-token instructions for direct generation of professional infographics including PPTs posters comics and more. Stronger Semantic Adherence: Native 2K resolution support for finely detailed realistic scenes including people nature and architecture. Improved Text Rendering: Integrated understanding and generation capabilities unifying image generation and editing in a single mode Lighter Model Architecture:"
X Link 2026-02-10T14:26Z 18.6K followers, 99.5K engagements

"3/4 Blog: Try it now https://chat.qwen.ai/inputFeature=t2i https://qwen.ai/blogid=qwen-image-2.0 https://chat.qwen.ai/inputFeature=t2i https://qwen.ai/blogid=qwen-image-2.0"
X Link 2026-02-10T14:26Z 18.6K followers, [----] engagements

"4/4 Don't miss out on the discussion. Join the server now 👇 https://discord.com/invite/z9QeEveYB2 https://discord.com/invite/z9QeEveYB2"
X Link 2026-02-10T14:26Z 18.6K followers, [----] engagements

"Reimagining the life of Leonardo da Vinci through the lens of Z-Image-Turbo. 🎨 From his early sketches in Florence to the masterpieces that changed history weve brought the story of the ultimate polymath to life. Every frame in this video was generated by Z-Image-Turbo showcasing its ability to capture intricate historical details and timeless artistic styles. Witness the fusion of Renaissance genius and Z-Image-Turbo. https://twitter.com/i/web/status/2022146299123122285 https://twitter.com/i/web/status/2022146299123122285"
X Link 2026-02-13T03:10Z 18.6K followers, [----] engagements

"Hello creators and builders This week wasnt just about new models it was about new foundations.Were thrilled to unveil Qwen-Image-2.0 our next-generation image generation model. And for the first time were also open-sourcing Zvec the SQLite of vector databases.We are bringing accessible high-performance AI systems to the users not just models but full-stack infrastructure you can run anywhere. Lets dive in. https://open.substack.com/pub/tongyilab/p/feb-13-2026-the-tongyi-weeklyr=6te3ta&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true"
X Link 2026-02-13T08:38Z 18.6K followers, 16.4K engagements

"1/2 Qwen3.5 is here. The next frontier of Native Multimodal Agents is open. 🚀 We are thrilled to releaseQwen3.5-397B-A17B our flagship open-weight vision-language model. Built for the future of coding reasoning and seamless multimodal interaction. Key Highlights: Inference Efficiency: A massive 397B total parameters but only17B activedelivering flagship power at a fraction of the cost. Hybrid Architecture: Innovative Gated Delta Networks (Linear Attention) + Sparse MoE for extreme speed. True Multimodality: Exceptional performance across GUI interaction video comprehension and agentic"
X Link 2026-02-16T09:59Z 18.6K followers, [----] engagements

"2/2 Explore the weights & report: Blog: ModelScope: Hugging Face: https://huggingface.co/collections/Qwen/qwen35 https://qwen.ai/blogid=qwen3.5 https://huggingface.co/collections/Qwen/qwen35 https://qwen.ai/blogid=qwen3.5"
X Link 2026-02-16T09:59Z 18.6K followers, [---] engagements

"Interested in seeing the clean line art behind your favorite anime characters @Yeq6X developed a new LoRA using Qwen-Image-Edit-2509 that effectively extracts sketches from existing images. Its a handy tool for anyone wanting to study character structure. Give it a try and see how it works with your images. https://twitter.com/i/web/status/2011716013537038844 https://twitter.com/i/web/status/2011716013537038844"
X Link 2026-01-15T08:24Z 18K followers, [----] engagements

"Look at those silk textures. Qwen-Image-2512 dodged the "AI sheen" entirely delivering hyper-realistic fabric and lighting that actually makes sense. Prompt is tucked in the Altgo give Qwen-Image-2512 a spin and see the magic for yourself"
X Link 2026-01-15T09:19Z 18K followers, [----] engagements

"For those using ComfyUI here is an interesting new workflow leveraging Z-Image. It provides a more refined sampling method for those who prioritize high-fidelity textures and intricate designs in their work. We really appreciate the developer's contribution to expanding whats possible with the model. https://twitter.com/i/web/status/2011734030660604211 https://twitter.com/i/web/status/2011734030660604211"
X Link 2026-01-15T09:35Z 18K followers, [----] engagements

"Its always inspiring to see how the community pushes our models in new directions. A developer has released LightingRemap Alpha [---] a LoRA specifically designed to refine lighting effects using Qwen-Image-Edit-2511. It provides a more precise way to manage highlights and shadows giving your edits a more professional finish. Grateful to the dev for sharing this with everyone. https://twitter.com/i/web/status/2012051483429388430 https://twitter.com/i/web/status/2012051483429388430"
X Link 2026-01-16T06:37Z 18K followers, [----] engagements

"For those looking into AI-driven interior design this workflow is a great resource. By leveraging Qwen-Image-Edit-2511 the developer demonstrates how to transform 2D furniture images into 3D-aware placements. This allows for realistic shadows and consistent textures making virtual staging more accessible and high-fidelity. We appreciate the developer sharing this innovative use case. Give it a try and see how it changes your staging process. Qwen-Image-Edit [----] : Virtually Staging Any Furniture in 3D - 16gb vram @Ali_TongyiLab @ComfyUI Youtube tutorial : https://t.co/wUScQHJ1Z6"
X Link 2026-01-16T09:13Z 18K followers, [----] engagements

"Hello creators and builders While this week didnt bring new model releases or research papers from our lab it was anything but quiet. In fact it was a brilliant showcase of community ingenuity developers building refining and reimagining whats possible with AIGC ecosystem. Lets celebrate what you built this week. https://open.substack.com/pub/tongyilab/p/jan-16-2026-the-tongyi-weeklyr=6te3ta&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true https://open.substack.com/pub/tongyilab/p/jan-16-2026-the-tongyi-weeklyr=6te3ta&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true"
X Link 2026-01-16T09:30Z 18K followers, [----] engagements

"Huge shoutout to @aleenaamiir for the spark. Were genuinely impressed by how Qwen-Image-2512 nailed the spatial depth herelook at that sofa perspective and the flawless window alignment. No digital hallucinations just clean matte textures. Prompts in the tweet quoted below. Go run it on Qwen-Image-2512. Clay-Style Soft Minimal 3D Illustrations Gemini Nano Banana Pro Prompt: Illustrate SUBJECT in a minimal clay-style 3D look with soft rounded forms matte textures and pastel tones. Keep the composition simple and playful avoiding realistic detailing. The result should feel"
X Link 2026-01-22T08:37Z 18K followers, [----] engagements

"Qwen3-TTS is now live on MLX-Audio. 🎧 Give it a spin and share your first audio samples with uswed love to hear what you create Qwen3-TTS now on MLX-Audio 🚀 uv pip install -U mlx-audio --prerelease=allow https://t.co/hDNlDZJnBg Qwen3-TTS now on MLX-Audio 🚀 uv pip install -U mlx-audio --prerelease=allow https://t.co/hDNlDZJnBg"
X Link 2026-01-23T03:36Z 18K followers, [----] engagements

"Huge shoutout to @Machinedelusion for sharing this incredible Qwen3-TTS workflow If youre looking to level up your local audio generation check out the full setup in the original post Update on the Qwen-TTS. Got the trainer working in full. However when the learning rate is too hot the grads explode and the model dies. It's really touchy but it is working technically. I'm gonna post it up as an experimental thing in the node pack for you guys to play https://t.co/iiH33Vtd6Y Update on the Qwen-TTS. Got the trainer working in full. However when the learning rate is too hot the grads explode and"
X Link 2026-01-26T10:08Z 18K followers, [----] engagements

"Tired of your product shots being stuck on a boring white background 🛋 This amazing "White_to_Scene" LoRA by @dx8152 is the solution Built on Qwen-Image-Edit-2509 it's an Image Fusion tool that lets you take any object and seamlessly place it into a brand new scene. Just use the trigger word "change white background to scene" and describe the scene you want. Go check out the tutorial and try it Link: https://huggingface.co/dx8152/Qwen-Image-Edit-2509-White_to_Scene https://huggingface.co/dx8152/Qwen-Image-Edit-2509-White_to_Scene"
X Link 2025-11-12T07:54Z 18.6K followers, 20.3K engagements

"4/4 more cases are below👇"
X Link 2025-08-28T09:15Z 18.6K followers, [----] engagements

"New toy for Qwen-Image users 🛠 Check out "Boreal" by kudzueye. It's an experimental LoRA designed for realistic photography. There's a ComfyUI workflow included to get you started. Link: https://huggingface.co/kudzueye/boreal-qwen-image https://huggingface.co/kudzueye/boreal-qwen-image"
X Link 2025-11-19T03:40Z 18.4K followers, 15.1K engagements

"Qwen-VL decodes the details. Qwen-Image sets the style. 🎨 DiffSynth-Studio has built a powerful suite by integrating our vision-language reasoning with our image generation capabilities. This is exactly the kind of full-stack innovation we love to see in the open-source community. 🔥BREAKING DiffSynth-Studio just dropped: Qwen-Image-i2L the first open-source suite that turns a single image into a custom LoRA 🖼➡🧠 Choose your flavor: 🎨 Style pure aesthetic extraction (2.4B) 🧩 Coarse captures content + style (7.9B) ✨ Fine 1024x1024 detail https://t.co/w4xF4Nf9nX 🔥BREAKING DiffSynth-Studio"
X Link 2025-12-09T16:09Z 18.1K followers, 12.3K engagements

"We just used this prompt to generate classical scenes from the Four Great Classical Novels at Z-Image-Turbo and the results are stunning. Its actually capturing the vibe of ancient Chinese literatureJourney to the West Dream of the Red Chamber all of it. The cultural understanding is deep here. Go test the model yourself. Nano Banana Pro https://t.co/0JfyqrAfJj Nano Banana Pro https://t.co/0JfyqrAfJj"
X Link 2025-12-10T04:20Z 18.1K followers, 35.6K engagements

"Seriously impressive work on this fan poster. Z-Image really shined on the third "Peach Garden Oath" generation nailing the specific mood and cultural details of the scene. Thanks for sharing this masterpiece with us. prompt/3D/ https://t.co/0bMq7IY9Vf prompt/3D/ https://t.co/0bMq7IY9Vf"
X Link 2025-12-10T12:57Z 18.1K followers, [----] engagements

"Hello builders and researchers This week was nothing short of extraordinary for Qwen a true harvest of research milestones product breakthroughs and community-powered innovation. From multilingual TTS that sounds human to RL methods that train smarter were witnessing the full arc of what open thoughtful AI can become. Lets dive in. https://open.substack.com/pub/tongyilab/p/dec-12-2025-the-tongyi-weeklyr=6te3ta&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true"
X Link 2025-12-12T05:40Z 18.1K followers, [----] engagements

"1/4 Were releasing MAI-UIa family of foundation GUI agents. It natively integrates MCP tool use agent user interaction devicecloud collaboration and online RL establishing state-of-the-art results in general GUI grounding and mobile GUI navigation surpassing Gemini-2.5-Pro Seed1.8 and UI-Tars-2 on AndroidWorld. To meet real-world deployment constrains MAI-UI includes a full-spectrum of sizes including 2B 8B 32B and 235B-A22B variants. We are publicly releasing two models: MAI-UI-2B and MAI-UI-8B"
X Link 2025-12-29T23:52Z 18K followers, 857.5K engagements

"2/4 Performance: MAI-UI establishes new state-of-the-art across GUI grounding and mobile navigation. It reaches 73.5% on ScreenSpot-Pro 91.3% on MMBench GUI L2 70.9% on OSWorld-G and 49.2% on UI-Vision surpassing Gemini-3-Pro and Seed1.8 on ScreenSpot-Pro. On mobile GUI navigation it sets a new SOTA of 76.7% on AndroidWorld surpassing UI-Tars-2 Gemini-2.5-Pro and Seed1.8. On MobileWorld MAI-UI significantly outperforms end-to-end GUI models. https://twitter.com/i/web/status/2005789042072506641 https://twitter.com/i/web/status/2005789042072506641"
X Link 2025-12-29T23:52Z 18K followers, [----] engagements

"MobileWorld: Pushing the Boundaries of Future Smartphone Interaction The latest breakthrough from the MAI Team at Tongyi Lab As a next-generation benchmark for mobile-agent evaluation MobileWorld transcends the limitations of traditional ones by realistically simulating users complex real-world demands. https://twitter.com/i/web/status/2005847750370549984 https://twitter.com/i/web/status/2005847750370549984"
X Link 2025-12-30T03:45Z 18.1K followers, 207.9K engagements

"Evaluation results reveal a stark reality: even the current state-of-the-art (SOTA) models achieve only a 51.7% success rate with end-to-end models peaking at just 20.9%. On the agent-user interaction and MCP-argumented tasks mainstream agents success rate drops nearly to zero highlighting a significant gap between agent's capabilities and real-world deployment readiness. MobileWorld is more than just an upgraded benchmarkits a critical stepping stone toward real-world mobile intelligence"
X Link 2025-12-30T03:45Z 18.1K followers, [----] engagements

"The codebase is now open-sourceresearchers worldwide are invited to push the limits GitHub: Official Website: arXiv: https://arxiv.org/abs/2512.19432 https://tongyi-mai.github.io/MobileWorld/ https://github.com/Tongyi-MAI/MobileWorld https://arxiv.org/abs/2512.19432 https://tongyi-mai.github.io/MobileWorld/ https://github.com/Tongyi-MAI/MobileWorld"
X Link 2025-12-30T03:45Z 18.1K followers, [----] engagements

"You asked Qwen-Image-Layered delivered. DiffSynth-Studio has open-sourced Qwen-Image-Layered-Control 🚀 It enables precise text-controlled layer extractionperfect for poster decomposition and graphic design workflows. Control image layers like never before: ✅ Single-layer output via text prompts ✅ Native https://t.co/4QwOKqTFLs DiffSynth-Studio has open-sourced Qwen-Image-Layered-Control 🚀 It enables precise text-controlled layer extractionperfect for poster decomposition and graphic design workflows. Control image layers like never before: ✅ Single-layer output via text prompts ✅ Native"
X Link 2026-01-13T08:20Z 18.1K followers, [----] engagements

"Qwen-Image-Edit-2511 is basically a playground right now. Were curiousif you were going to build the most needed adapter for this model today which lane would you lean into 🙌 Cast your vote below and let us know where the energy is. Semantic Reasoning Spatial & Physical Logic Other (Plz cmt below) Inpainting Fidelity Semantic Reasoning Spatial & Physical Logic Other (Plz cmt below) Inpainting Fidelity"
X Link 2026-01-13T09:37Z 18.3K followers, [----] engagements

"Grab it here https://huggingface.co/dx8152/Qwen-Image-Edit-2511-Gaussian-Splash https://huggingface.co/dx8152/Qwen-Image-Edit-2511-Gaussian-Splash"
X Link 2026-01-14T02:10Z 18.1K followers, [----] engagements

"Art meets AI in the best way possible Massive thanks to @Yeq6X for this Qwen-Image-Edit-2509 masterpiece. A beautiful reminder of how talented our creators are 👏 🎨Qwen Image Edit AI https://t.co/6lUSFmTbMl https://t.co/58ANxLysyR 🎨Qwen Image Edit AI https://t.co/6lUSFmTbMl https://t.co/58ANxLysyR"
X Link 2026-01-14T06:45Z 18.1K followers, [----] engagements

"QT your favorite sport sculpture using Qwen-Image-2512. Prompt in ALT"
X Link 2026-01-16T11:08Z 18.1K followers, [----] engagements

"Its great to see the community developing specialized tools on top of Qwen-Image-Edit-2511. This Object Remover LoRA provides a more focused approach to decluttering images ensuring that the underlying textures remain consistent after an object is gone. Its a thoughtful implementation that makes complex editing tasks a bit more manageable. We appreciate the @prithivMLmods's contribution to the community. https://twitter.com/i/web/status/2013089823813210241 https://twitter.com/i/web/status/2013089823813210241"
X Link 2026-01-19T03:23Z 18K followers, [----] engagements

"Explore the LoRA here: Wed love to see what you can achieve with it. https://huggingface.co/prithivMLmods/Qwen-Image-Edit-2511-Object-Remover https://huggingface.co/prithivMLmods/Qwen-Image-Edit-2511-Object-Remover"
X Link 2026-01-19T03:23Z 18.1K followers, [----] engagements

"Just look at this Hangzhou piece. The prompt called for a "pure white monochromatic clay render" with "high visual density" and Qwen-Image-2512 absolutely delivered. @0x00_Krypt thanks for sharing 3D Banana Prompt👇 --- A highly detailed 3D bas-relief poster design representing CITY_NAME. A dense collage composition featuring the most iconic landmarks and skyline of CITY_NAME. Foreground filled with local transportation systems busy highway networks with https://t.co/g7C8z4MJ8j 3D Banana Prompt👇 --- A highly detailed 3D bas-relief poster design representing CITY_NAME. A dense collage"
X Link 2026-01-19T03:45Z 18.1K followers, [----] engagements

"👀 Take a look at this new creation over on ModelScope. Real-Qwen-Image-V2 is here 📸 Developed by wikeeyang this fine-tuned version of Qwen-Image-2512 is built for those who demand peak realism and production-grade quality. 🤖 Model: https://t.co/XBA77J9gTz Key Highlights: ✨ Sharper details & enhanced realism ✨ Optimized for https://t.co/fa1kye97ri Real-Qwen-Image-V2 is here 📸 Developed by wikeeyang this fine-tuned version of Qwen-Image-2512 is built for those who demand peak realism and production-grade quality. 🤖 Model: https://t.co/XBA77J9gTz Key Highlights: ✨ Sharper details & enhanced"
X Link 2026-01-19T04:15Z 18.1K followers, [----] engagements

"Qwen-Image-2512 nailed the Guochao brief💅 It captures that "New Chinese" mood without trying too hard. Check the Alt for the prompt. QT Your Dragon https://t.co/qDXtQDtk5M QT Your Dragon https://t.co/qDXtQDtk5M"
X Link 2026-01-19T07:21Z 18.1K followers, [----] engagements

"This new workflow by Martin Rizzo brings incredible versatility to Z-Image-Turbo users. Its standout feature is a dedicated Style Selector with [--] customizable presets allowing you to switch artistic directions instantly. Beyond styling the integrated Refiner and Upscaler ensure that every output reaches a professional level of detail and resolution. A big thank you to the developer for building such a comprehensive toolkit. https://twitter.com/i/web/status/2013521323599810716 https://twitter.com/i/web/status/2013521323599810716"
X Link 2026-01-20T07:57Z 18.1K followers, 10.1K engagements

"You can find the workflow here: Give it a try and see how these styles transform your renders https://github.com/martin-rizzo/AmazingZImageWorkflow/tree/master https://github.com/martin-rizzo/AmazingZImageWorkflow/tree/master"
X Link 2026-01-20T07:57Z 18.1K followers, [----] engagements

"The community creativity truly makes our open-source models shine ✨ This stunning video showcases the seamless integration of Wan [---] Z-Image-Turbo and Qwen-Image-Edit-2511 to achieve a truly cinematic result. Its a powerful demonstration of how combining our open-source models empowers creators to produce truly exceptional work. A massive thank you to the creator for this masterpiece. Watch the full video here: https://www.youtube.com/watchv=54IxX6FtKg8 https://www.youtube.com/watchv=54IxX6FtKg8"
X Link 2026-01-21T06:05Z 18.1K followers, [----] engagements

"This toolkit solves common pain points by offering one-click style switching and high-speed "Turbo" configurations (optimizing for as few as [--] steps). It effectively brings precision and ease-of-use to the Z-Image ecosystem. Huge thanks to Martin Rizzo for this contribution to the open-source community. https://twitter.com/i/web/status/2013894405506080915 https://twitter.com/i/web/status/2013894405506080915"
X Link 2026-01-21T08:40Z 18.1K followers, [----] engagements

"Check it out here: https://github.com/martin-rizzo/ComfyUI-ZImagePowerNodes/tree/master https://github.com/martin-rizzo/ComfyUI-ZImagePowerNodes/tree/master"
X Link 2026-01-21T08:40Z 18.1K followers, [----] engagements

"Were thrilled to open-source the entire Qwen3-TTS series It's the ultimate toolkit for voice cloning creative "Voice Design" and ultra-realistic human speech generated via natural language prompts. Core Highlights: -Ultra-Low Latency: Our innovative Dual-Track modeling delivers the first audio packet after just ONE character. End-to-end latency as low as 97ms -Global Reach: Full support for [--] major languages (CN EN JP KR DE FR RU PT ES IT) and diverse dialects. -Efficient Power: High-fidelity non-DiT architecture available in two sizes: 1.7B (Peak Performance) & 0.6B (High Efficiency). -SOTA"
X Link 2026-01-22T13:27Z 18.1K followers, 141K engagements

"Here is the link of ModelScope: https://modelscope.cn/collections/Qwen/Qwen3-TTS https://modelscope.cn/collections/Qwen/Qwen3-TTS"
X Link 2026-01-22T13:44Z 18.1K followers, [----] engagements

"@FurkanGozukara of course its good 😊"
X Link 2026-01-22T13:45Z 18.1K followers, [----] engagements

"Don't miss out on the discussion. Join the server now 👇 https://discord.gg/z9QeEveYB2 https://discord.gg/z9QeEveYB2"
X Link 2026-01-23T05:51Z 18.1K followers, [----] engagements

"Hello creators and builders With the full Qwen3-TTS family now open-sourced this week anyone can now design clone or fine-tune voices with studio-grade quality in [--] languages using just 0.6B parameters. Lets dive in. https://open.substack.com/pub/tongyilab/p/jan-23-2026-the-tongyi-weeklyr=6te3ta&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true https://open.substack.com/pub/tongyilab/p/jan-23-2026-the-tongyi-weeklyr=6te3ta&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true"
X Link 2026-01-23T06:23Z 18K followers, [----] engagements

"Youll never guess how this ends. I'm so dope like yeah We gon' blow like🎈 yeah Made with @ComfyUI + @Alibaba_Wan Wan Animate + @Ali_TongyiLab Qwen Edit [----] 💥 https://t.co/sY5PgiE6rj I'm so dope like yeah We gon' blow like🎈 yeah Made with @ComfyUI + @Alibaba_Wan Wan Animate + @Ali_TongyiLab Qwen Edit [----] 💥 https://t.co/sY5PgiE6rj"
X Link 2026-01-26T02:01Z 18.1K followers, [----] engagements

"Heres the Qwen3-TTS technical report. Check it out when you have a chance https://arxiv.org/abs/2601.15621 https://arxiv.org/abs/2601.15621"
X Link 2026-01-26T02:09Z 18.1K followers, 10.4K engagements

"Community creativity is on another levelwere blown away by NuMarkdown-8B-Thinking a brilliant fine-tune of Qwen 2.5-VL-7B that brings genuine reasoning to OCR. Extracts text from images instantly. 99% OCR accuracy. Production-ready document digitization. Zero manual typing. https://t.co/VI83N6ff44 Extracts text from images instantly. 99% OCR accuracy. Production-ready document digitization. Zero manual typing. https://t.co/VI83N6ff44"
X Link 2026-01-26T03:06Z 18.1K followers, [----] engagements

"TheSuperior-Reasoning-SFT-gpt-oss-120bdataset has been released. It's a high-quality open-source dataset of 435K carefully curated samples specifically designed to democratize the training of high-performance Long Chain-of-Thought models. It has held the #1 position on Hugging Faces Trending Datasets list since January [--] and surpassed 13K downloads. Here is the link of dataset: https://huggingface.co/datasets/Alibaba-Apsara/Superior-Reasoning-SFT-gpt-oss-120b https://huggingface.co/datasets/Alibaba-Apsara/Superior-Reasoning-SFT-gpt-oss-120b"
X Link 2026-01-26T03:13Z 18.3K followers, [----] engagements

"DASD-4B-Thinking has demonstrated state-of-the-art reasoning performance on AIME LCB and GPQA owing to our proposed Distribution-Aligned Sequence Distillation (DASD) pipeline which tackles two core challenges: distributional mismatch between teacher and student model outputs and exposure bias during sequential generation. Total downloads for the DASD-4B-Thinking model and its derivatives have exceeded [-----] times on Hugging Face. ModelScope: Hugging Face: https://huggingface.co/Alibaba-Apsara/DASD-4B-Thinking https://modelscope.cn/models/Alibaba-Apsara/DASD-4B-Thinking"
X Link 2026-01-26T03:13Z 18.1K followers, [----] engagements

"Curious how Apsara-Reason-4B achieved such outstanding results These are our secrets"
X Link 2026-01-26T03:13Z 18.1K followers, [----] engagements

"We are thrilled to launch Qwen3-Max-Thinking the new flagship of the Qwen family. Pre-trained on a massive 36T token dataset this model represents a paradigm shift in how LLMs approach complex problem-solving. Why Qwen3-Max-Thinking stands out: Smart Test-Time Scaling: We moved beyond standard parallel sampling. Our new experience-cumulative strategy allows the model to learn from previous reasoning steps within the same context. This iterative self-reflection boosts efficiency and intelligence achieving a record-breaking [----] on the HLEsurpassing Gemini [--] Pro and GPT-5.2-Thinking. Native"
X Link 2026-01-26T15:25Z 18.2K followers, 216.1K engagements

"Performance: Across [--] major benchmarks covering coding math and instruction following Qwen3-Max-Thinking has set new SOTA records including perfect scores on AIME [--] and HMMT 25"
X Link 2026-01-26T15:25Z 18.2K followers, [----] engagements

"Availability: Developers and researchers can access the model today via the Alibaba Cloud Bailian API or experience the capabilities firsthand on QwenChat. 🔗 Try it here: https://chat.qwen.ai/ https://chat.qwen.ai/"
X Link 2026-01-26T15:25Z 18.2K followers, [----] engagements

"Don't miss out on the discussion. Join the server now https://discord.gg/z9QeEveYB2 https://discord.gg/z9QeEveYB2"
X Link 2026-01-26T15:25Z 18K followers, [----] engagements

"Want to know how to use Qwen3-Max-Thinking on Qwen Chat Watch this @cherry_cc12 thank you for this fantastic demo and for your long-standing support. Inspired by @JustinLin610 's suggestion "Why not build a demo to show everyone how to use Qwen3-Max-Thinking on Qwen Chat" this little demo is here. Dedicated to the nearly three wonderful years of walking alongside Qwen 🌟 https://t.co/N70VNOjAIq Inspired by @JustinLin610 's suggestion "Why not build a demo to show everyone how to use Qwen3-Max-Thinking on Qwen Chat" this little demo is here. Dedicated to the nearly three wonderful years of"
X Link 2026-01-27T03:01Z 18K followers, [----] engagements

"(_) ⚡Eyes open. ) Stay tuned. /"
X Link 2026-01-27T06:38Z 18.3K followers, 35.2K engagements

"@xeophon were following UTC+8 time for this"
X Link 2026-01-27T06:41Z 18.1K followers, [----] engagements

"@1amInvesting Were following UTC+8 time for this. You might wake up to a nice surprise tomorrow morning"
X Link 2026-01-27T06:54Z 18.1K followers, [----] engagements

"@francis_cgl The surprise weve all been waiting for"
X Link 2026-01-27T07:14Z 18.1K followers, [----] engagements

"2/6 🎨 Aesthetic & Artistic Diversity Z-Image masters a vast spectrum of visual languagesfrom hyper-realistic photography and cinematic digital art to intricate anime and stylized illustrations. It is the ideal engine for scenarios requiring rich multi-dimensional expression"
X Link 2026-01-27T16:28Z 18.3K followers, [----] engagements

"3/6🛠 Undistilled Foundation As a non-distilled base model Z-Image preserves the complete training signal. It supports full Classifier-Free Guidance (CFG) providing the precision required for complex prompt engineering and professional workflows. Built for Development The ideal starting point for the community. Its non-distilled nature makes it a good base for LoRA training structural conditioning (ControlNet) and semantic conditioning. https://twitter.com/i/web/status/2016186686103761148 https://twitter.com/i/web/status/2016186686103761148"
X Link 2026-01-27T16:28Z 18.3K followers, [----] engagements

"4/6🧬 Enhanced Output Diversity Built for exploration Z-Image delivers significantly higher variability in composition facial identity and lighting across different seeds ensuring that multi-person scenes remain distinct and dynamic"
X Link 2026-01-27T16:29Z 18.1K followers, [----] engagements

"5/6 🚫 Robust Negative Control Responds with high fidelity to negative prompting allowing users to reliably suppress artifacts and adjust compositions"
X Link 2026-01-27T16:29Z 18.1K followers, [----] engagements

"6/6 We invite the community's active participation and feedback to help us build a generative AI ecosystem that is not only open and transparent but also more efficient accessible and sustainable. GitHub: ModelScope: HuggingFace: Z-Image gallry : https://www.modelscope.cn/studios/Tongyi-MAI/Z-Image-Gallery https://huggingface.co/Tongyi-MAI/Z-Image https://modelscope.cn/models/Tongyi-MAI/Z-Image https://github.com/Tongyi-MAI/Z-Image https://modelscope.cn/models/Tongyi-MAI/Z-Image https://github.com/Tongyi-MAI/Z-Image https://www.modelscope.cn/studios/Tongyi-MAI/Z-Image-Gallery"
X Link 2026-01-27T16:29Z 18.1K followers, [----] engagements

"Don't miss out on the discussion. Join the server now https://discord.gg/z9QeEveYB2 https://discord.gg/z9QeEveYB2"
X Link 2026-01-27T16:34Z 18.1K followers, [----] engagements

"😗 Z-Image is natively supported in ComfyUI on Day [--] The non-distilled Z-Image base model is a true foundation model ideal for fine-tuning customization and community-driven development. - Diverse aesthetics: from photorealism to expressive styles - High generation diversity: https://t.co/luR6gcVgt7 Z-Image is natively supported in ComfyUI on Day [--] The non-distilled Z-Image base model is a true foundation model ideal for fine-tuning customization and community-driven development. - Diverse aesthetics: from photorealism to expressive styles - High generation diversity: https://t.co/luR6gcVgt7"
X Link 2026-01-28T02:14Z 18.1K followers, [----] engagements

"🥰 Free training for Z-Image is now live on our international site Open to everyonego check it out https://t.co/tN0Ej2MvB4 👉Any question Join our Discord: https://t.co/7wDmegDrla https://t.co/QKme8b5XWk Free training for Z-Image is now live on our international site Open to everyonego check it out https://t.co/tN0Ej2MvB4 👉Any question Join our Discord: https://t.co/7wDmegDrla https://t.co/QKme8b5XWk"
X Link 2026-01-28T03:05Z 18.1K followers, [----] engagements

"When Monet meets AI. Created with Z-Image & Wan2.2. QT your Artworks"
X Link 2026-01-28T10:09Z 18.2K followers, [----] engagements

"Hello developers This week was a landmark release week. The long-awaited Z-Image foundation model is now live bringing full-capacity undistilled power to visual creation. Qwen3-Max-Thinking our new flagship reasoning model has arrived with strong performance. And Qwen3-ASR our open-source speech recognition suite redefines accuracy in noisy multilingual real-world settings. Lets dive in. https://open.substack.com/pub/tongyilab/p/jan-30-2026-the-tongyi-weeklyr=6te3ta&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true"
X Link 2026-01-30T08:16Z 18.3K followers, [----] engagements

"2/6 Choose Your Path. Define the Impact.🚀 Track 1: AI for Production.Build LoRAs for branding e-commerce and industrial design. Transform VI elements into professional-grade assets. Track 2: AI for Good.Use AI to preserve cultural heritage or build accessibility tools like tactile image generators for the visually impaired. How will you innovate"
X Link 2026-02-02T10:12Z 18.2K followers, [---] engagements

"3/6 High-Performance Creativity High-Stakes Rewards. 🏆We are looking for quality innovation and industry potential.Our prize pool includes: 🥇Gold:iPhone [--] Pro Max (512G) 🥈Silver:Sony PS5 Pro 🥉Bronze:$800 Gift Cards Evaluation:Visual Quality (30%) Innovation (25%) Potential (25%) and Community Influence (20%). https://twitter.com/i/web/status/2018266386196463686 https://twitter.com/i/web/status/2018266386196463686"
X Link 2026-02-02T10:12Z 18.2K followers, [---] engagements

"2/6 Choose Your Path. Define the Impact.🚀 Track 1: AI for Production.Build LoRAs for branding e-commerce and industrial design. Transform VI elements into professional-grade assets. Track 2: AI for Good.Use AI to preserve cultural heritage or build accessibility tools like tactile image generators for the visually impaired. How will you innovate"
X Link 2026-02-02T11:00Z 18.3K followers, [----] engagements

"3/6 High-Performance Creativity High-Stakes Rewards. 🏆We are looking for quality innovation and industry potential.Our prize pool includes: 🥇Gold:iPhone [--] Pro Max (512G) 🥈Silver:Sony PS5 Pro 🥉Bronze:$800 Gift Cards Evaluation:Visual Quality (30%) Innovation (25%) Potential (25%) and Community Influence (20%). https://twitter.com/i/web/status/2018278430652186975 https://twitter.com/i/web/status/2018278430652186975"
X Link 2026-02-02T11:00Z 18.3K followers, [----] engagements

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