Adina Yakup
@adinayakup.bsky.social
AI Research @Hugging Face 🤗
Contributing to the Chinese ML community.
Contributing to the Chinese ML community.
✨ Supports 33 languages, including 5 ethnic minority languages in China 👀
✨ Including a translation ensemble model: Chimera-7B
✨ Full pipeline: pretrain > CPT > SFT > enhancement > ensemble refinement > SOTA performance at similar scale
✨ Including a translation ensemble model: Chimera-7B
✨ Full pipeline: pretrain > CPT > SFT > enhancement > ensemble refinement > SOTA performance at similar scale
September 1, 2025 at 9:15 AM
✨ Supports 33 languages, including 5 ethnic minority languages in China 👀
✨ Including a translation ensemble model: Chimera-7B
✨ Full pipeline: pretrain > CPT > SFT > enhancement > ensemble refinement > SOTA performance at similar scale
✨ Including a translation ensemble model: Chimera-7B
✨ Full pipeline: pretrain > CPT > SFT > enhancement > ensemble refinement > SOTA performance at similar scale
✨ 560B total / ~27B active MoE — MIT license
✨ 128k context length + advanced reasoning
✨ ScMoE design: 100+ TPS inference
✨ Stable large-scale training + strong agentic performance
✨ 128k context length + advanced reasoning
✨ ScMoE design: 100+ TPS inference
✨ Stable large-scale training + strong agentic performance
September 1, 2025 at 8:26 AM
✨ 560B total / ~27B active MoE — MIT license
✨ 128k context length + advanced reasoning
✨ ScMoE design: 100+ TPS inference
✨ Stable large-scale training + strong agentic performance
✨ 128k context length + advanced reasoning
✨ ScMoE design: 100+ TPS inference
✨ Stable large-scale training + strong agentic performance
✨ Large-scale triplet dataset (content, style, stylized)
✨ Disentangled learning: style alignment + content preservation
✨ Style Reward Learning (SRL) for higher fidelity
✨ USO-Bench: 1st benchmark for style & subject jointly
✨ SOTA results on subject consistency & style similarity
✨ Disentangled learning: style alignment + content preservation
✨ Style Reward Learning (SRL) for higher fidelity
✨ USO-Bench: 1st benchmark for style & subject jointly
✨ SOTA results on subject consistency & style similarity
August 29, 2025 at 9:18 AM
✨ Large-scale triplet dataset (content, style, stylized)
✨ Disentangled learning: style alignment + content preservation
✨ Style Reward Learning (SRL) for higher fidelity
✨ USO-Bench: 1st benchmark for style & subject jointly
✨ SOTA results on subject consistency & style similarity
✨ Disentangled learning: style alignment + content preservation
✨ Style Reward Learning (SRL) for higher fidelity
✨ USO-Bench: 1st benchmark for style & subject jointly
✨ SOTA results on subject consistency & style similarity
Wan2.2 🔥A video diffusion model with MoE just released by Alibaba_Wan
huggingface.co/Wan-AI/Wan2....
huggingface.co/Wan-AI/Wan2....
huggingface.co/Wan-AI/Wan2....
huggingface.co/Wan-AI/Wan2....
July 28, 2025 at 1:46 PM
Wan2.2 🔥A video diffusion model with MoE just released by Alibaba_Wan
huggingface.co/Wan-AI/Wan2....
huggingface.co/Wan-AI/Wan2....
huggingface.co/Wan-AI/Wan2....
huggingface.co/Wan-AI/Wan2....
GLM-4.5 🔥 The largest open models yet from Zhipu AI.
Built for intelligent agents with unified capabilities: reasoning, coding, tool use.
huggingface.co/collections/...
✨ 355B total / 32B active - MIT license
✨ Hybrid modes: Thinking mode for complex tasks/ Non-thinking mode for instant replies
Built for intelligent agents with unified capabilities: reasoning, coding, tool use.
huggingface.co/collections/...
✨ 355B total / 32B active - MIT license
✨ Hybrid modes: Thinking mode for complex tasks/ Non-thinking mode for instant replies
July 28, 2025 at 1:28 PM
GLM-4.5 🔥 The largest open models yet from Zhipu AI.
Built for intelligent agents with unified capabilities: reasoning, coding, tool use.
huggingface.co/collections/...
✨ 355B total / 32B active - MIT license
✨ Hybrid modes: Thinking mode for complex tasks/ Non-thinking mode for instant replies
Built for intelligent agents with unified capabilities: reasoning, coding, tool use.
huggingface.co/collections/...
✨ 355B total / 32B active - MIT license
✨ Hybrid modes: Thinking mode for complex tasks/ Non-thinking mode for instant replies
Qwen3-Coder 💻 agentic code model by Alibaba_Qwen
huggingface.co/collections/...
✨ 480B total, 35B activated MoE
✨ Agentic Coding + Browser Use → Top code model performance
✨ 256K context (up to 1M via Yarn) for repo-scale understanding
huggingface.co/collections/...
✨ 480B total, 35B activated MoE
✨ Agentic Coding + Browser Use → Top code model performance
✨ 256K context (up to 1M via Yarn) for repo-scale understanding
July 22, 2025 at 9:32 PM
Qwen3-Coder 💻 agentic code model by Alibaba_Qwen
huggingface.co/collections/...
✨ 480B total, 35B activated MoE
✨ Agentic Coding + Browser Use → Top code model performance
✨ 256K context (up to 1M via Yarn) for repo-scale understanding
huggingface.co/collections/...
✨ 480B total, 35B activated MoE
✨ Agentic Coding + Browser Use → Top code model performance
✨ 256K context (up to 1M via Yarn) for repo-scale understanding
RoboBrain 2.0🔥 OPEN embedded brain model by BAAIBeijing
huggingface.co/BAAI/RoboBra...
✨ 7B - Apache 2.0 / 32B coming soon
✨ Supports multiple images, long videos, and high-resolution visuals
✨ Spatial + temporal reasoning
✨ Real-time memory & scene graphs
huggingface.co/BAAI/RoboBra...
✨ 7B - Apache 2.0 / 32B coming soon
✨ Supports multiple images, long videos, and high-resolution visuals
✨ Spatial + temporal reasoning
✨ Real-time memory & scene graphs
June 9, 2025 at 10:09 AM
RoboBrain 2.0🔥 OPEN embedded brain model by BAAIBeijing
huggingface.co/BAAI/RoboBra...
✨ 7B - Apache 2.0 / 32B coming soon
✨ Supports multiple images, long videos, and high-resolution visuals
✨ Spatial + temporal reasoning
✨ Real-time memory & scene graphs
huggingface.co/BAAI/RoboBra...
✨ 7B - Apache 2.0 / 32B coming soon
✨ Supports multiple images, long videos, and high-resolution visuals
✨ Spatial + temporal reasoning
✨ Real-time memory & scene graphs
Look who just joined the Enterprise Hub on huggingface🎉
👉 huggingface.co/baidu
Earlier this year, Baidu announced they would release their open-source work in June. Looks like they’re ready！
Follow the account & stay tuned!
👉 huggingface.co/baidu
Earlier this year, Baidu announced they would release their open-source work in June. Looks like they’re ready！
Follow the account & stay tuned!
June 5, 2025 at 10:06 AM
Look who just joined the Enterprise Hub on huggingface🎉
👉 huggingface.co/baidu
Earlier this year, Baidu announced they would release their open-source work in June. Looks like they’re ready！
Follow the account & stay tuned!
👉 huggingface.co/baidu
Earlier this year, Baidu announced they would release their open-source work in June. Looks like they’re ready！
Follow the account & stay tuned!
HunyuanPortrait 🔥 video model by Tencent Hunyuan team.
huggingface.co/papers/2503....
huggingface.co/tencent/Huny...
✨Portrait animation from just one image + a video prompt
✨Diffusion-based, implicit motion control
✨Superior temporal consistency & detail
huggingface.co/papers/2503....
huggingface.co/tencent/Huny...
✨Portrait animation from just one image + a video prompt
✨Diffusion-based, implicit motion control
✨Superior temporal consistency & detail
May 27, 2025 at 2:21 PM
HunyuanPortrait 🔥 video model by Tencent Hunyuan team.
huggingface.co/papers/2503....
huggingface.co/tencent/Huny...
✨Portrait animation from just one image + a video prompt
✨Diffusion-based, implicit motion control
✨Superior temporal consistency & detail
huggingface.co/papers/2503....
huggingface.co/tencent/Huny...
✨Portrait animation from just one image + a video prompt
✨Diffusion-based, implicit motion control
✨Superior temporal consistency & detail
Index-AniSora 🎬 an open anime video model released by BiliBili
👉 huggingface.co/IndexTeam/In...
✨ Apache2.0
✨ Supports many 2D styles: anime, manga, VTubers, and more
✨ Fine control over characters and actions with smart masking
👉 huggingface.co/IndexTeam/In...
✨ Apache2.0
✨ Supports many 2D styles: anime, manga, VTubers, and more
✨ Fine control over characters and actions with smart masking
May 19, 2025 at 3:16 PM
Index-AniSora 🎬 an open anime video model released by BiliBili
👉 huggingface.co/IndexTeam/In...
✨ Apache2.0
✨ Supports many 2D styles: anime, manga, VTubers, and more
✨ Fine control over characters and actions with smart masking
👉 huggingface.co/IndexTeam/In...
✨ Apache2.0
✨ Supports many 2D styles: anime, manga, VTubers, and more
✨ Fine control over characters and actions with smart masking
Wan2.1-FLF2V🎥 a 14B start-end frame video generation model just released by Alibaba_Wan🔥
huggingface.co/Wan-AI/Wan2....
✨ Give it two images (start & end), it generates a smooth, high-quality video in between.
✨ Apache 2.0 licensed
✨ Built on DiT + Flow Matching
huggingface.co/Wan-AI/Wan2....
✨ Give it two images (start & end), it generates a smooth, high-quality video in between.
✨ Apache 2.0 licensed
✨ Built on DiT + Flow Matching
April 17, 2025 at 4:09 PM
Wan2.1-FLF2V🎥 a 14B start-end frame video generation model just released by Alibaba_Wan🔥
huggingface.co/Wan-AI/Wan2....
✨ Give it two images (start & end), it generates a smooth, high-quality video in between.
✨ Apache 2.0 licensed
✨ Built on DiT + Flow Matching
huggingface.co/Wan-AI/Wan2....
✨ Give it two images (start & end), it generates a smooth, high-quality video in between.
✨ Apache 2.0 licensed
✨ Built on DiT + Flow Matching
✨Minimal & transparent pipeline with standard tools
✨Standardized eval to track training & reflection
✨Open Code & Dataset
✨Standardized eval to track training & reflection
✨Open Code & Dataset
April 8, 2025 at 8:15 AM
✨Minimal & transparent pipeline with standard tools
✨Standardized eval to track training & reflection
✨Open Code & Dataset
✨Standardized eval to track training & reflection
✨Open Code & Dataset
The top trending model on @hf.co today is is a great example of open source diversity🔥We’ve got:
🌍 Models from all over the world
🔠 Supporting multiple languages
✨ Built by tech giants, startups, and communities
Love seeing innovation without borders, powered by the global AI community♥️
🌍 Models from all over the world
🔠 Supporting multiple languages
✨ Built by tech giants, startups, and communities
Love seeing innovation without borders, powered by the global AI community♥️
March 18, 2025 at 10:16 AM
The top trending model on @hf.co today is is a great example of open source diversity🔥We’ve got:
🌍 Models from all over the world
🔠 Supporting multiple languages
✨ Built by tech giants, startups, and communities
Love seeing innovation without borders, powered by the global AI community♥️
🌍 Models from all over the world
🔠 Supporting multiple languages
✨ Built by tech giants, startups, and communities
Love seeing innovation without borders, powered by the global AI community♥️
New 3D models from Tencent Hunyuan are now on @hf.co 🔥
✨ Hunyuan3D-2mv: multiview shape model for high quality generation
✨ Hunyuan3D-2mini: 0.6B lightweight model for efficient workflows
Model:
huggingface.co/tencent/Huny...
huggingface.co/tencent/Huny...
Demo: huggingface.co/spaces/tence...
✨ Hunyuan3D-2mv: multiview shape model for high quality generation
✨ Hunyuan3D-2mini: 0.6B lightweight model for efficient workflows
Model:
huggingface.co/tencent/Huny...
huggingface.co/tencent/Huny...
Demo: huggingface.co/spaces/tence...
March 18, 2025 at 9:15 AM
New 3D models from Tencent Hunyuan are now on @hf.co 🔥
✨ Hunyuan3D-2mv: multiview shape model for high quality generation
✨ Hunyuan3D-2mini: 0.6B lightweight model for efficient workflows
Model:
huggingface.co/tencent/Huny...
huggingface.co/tencent/Huny...
Demo: huggingface.co/spaces/tence...
✨ Hunyuan3D-2mv: multiview shape model for high quality generation
✨ Hunyuan3D-2mini: 0.6B lightweight model for efficient workflows
Model:
huggingface.co/tencent/Huny...
huggingface.co/tencent/Huny...
Demo: huggingface.co/spaces/tence...
R1-Omni🔥RLVR-Powered Multimodal LLM released by Alibaba
Model: huggingface.co/StarJiaxing/...
Paper: huggingface.co/papers/2503....
✨0.5B with Apache2.0
✨ Improve emotion recognition with visual and audio cues
Model: huggingface.co/StarJiaxing/...
Paper: huggingface.co/papers/2503....
✨0.5B with Apache2.0
✨ Improve emotion recognition with visual and audio cues
March 12, 2025 at 10:39 AM
R1-Omni🔥RLVR-Powered Multimodal LLM released by Alibaba
Model: huggingface.co/StarJiaxing/...
Paper: huggingface.co/papers/2503....
✨0.5B with Apache2.0
✨ Improve emotion recognition with visual and audio cues
Model: huggingface.co/StarJiaxing/...
Paper: huggingface.co/papers/2503....
✨0.5B with Apache2.0
✨ Improve emotion recognition with visual and audio cues
CogView-4 is out🔥🚀 The SoTa OPEN text to image model by ZhipuAI
huggingface.co/THUDM/CogVie...
huggingface.co/spaces/THUDM...
✨ 6B with Apache2.0
✨ Supports Chinese & English Prompts by ANY length
✨ Generate Chinese characters within images
✨ Creates images at any resolution within a given range
huggingface.co/THUDM/CogVie...
huggingface.co/spaces/THUDM...
✨ 6B with Apache2.0
✨ Supports Chinese & English Prompts by ANY length
✨ Generate Chinese characters within images
✨ Creates images at any resolution within a given range
March 4, 2025 at 9:01 AM
CogView-4 is out🔥🚀 The SoTa OPEN text to image model by ZhipuAI
huggingface.co/THUDM/CogVie...
huggingface.co/spaces/THUDM...
✨ 6B with Apache2.0
✨ Supports Chinese & English Prompts by ANY length
✨ Generate Chinese characters within images
✨ Creates images at any resolution within a given range
huggingface.co/THUDM/CogVie...
huggingface.co/spaces/THUDM...
✨ 6B with Apache2.0
✨ Supports Chinese & English Prompts by ANY length
✨ Generate Chinese characters within images
✨ Creates images at any resolution within a given range
Snack Huggy is all set , and so are we!
Happy new year to everyone celebrating!! 🧧🐍🤗
Happy new year to everyone celebrating!! 🧧🐍🤗
January 29, 2025 at 9:39 AM
Snack Huggy is all set , and so are we!
Happy new year to everyone celebrating!! 🧧🐍🤗
Happy new year to everyone celebrating!! 🧧🐍🤗
What happened yesterday in the Chinese AI community? 🚀
huggingface.co/posts/AdinaY...
huggingface.co/posts/AdinaY...
January 21, 2025 at 11:51 AM
What happened yesterday in the Chinese AI community? 🚀
huggingface.co/posts/AdinaY...
huggingface.co/posts/AdinaY...
✨ MIT License : enabling distillation for custom models
✨ 32B & 70B models match OpenAI o1-mini in multiple capabilities
✨ API live now! Access Chain of Thought reasoning with model='deepseek-reasoner'
✨ 32B & 70B models match OpenAI o1-mini in multiple capabilities
✨ API live now! Access Chain of Thought reasoning with model='deepseek-reasoner'
January 20, 2025 at 12:44 PM
✨ MIT License : enabling distillation for custom models
✨ 32B & 70B models match OpenAI o1-mini in multiple capabilities
✨ API live now! Access Chain of Thought reasoning with model='deepseek-reasoner'
✨ 32B & 70B models match OpenAI o1-mini in multiple capabilities
✨ API live now! Access Chain of Thought reasoning with model='deepseek-reasoner'