Hugging Face (Twitter)
RT @maximelabonne: Liquid just released two 450M and 1.6B param VLMs!
They're super fast and leverage SigLIP2 NaFlex encoders to handle native resolutions without distortion.
Available today on @huggingface!
RT @maximelabonne: Liquid just released two 450M and 1.6B param VLMs!
They're super fast and leverage SigLIP2 NaFlex encoders to handle native resolutions without distortion.
Available today on @huggingface!
Hugging Face (Twitter)
RT @ramin_m_h: meet LFM2-VL: an efficient Liquid vision-language model for the device class. open weights, 440M & 1.6B, up to 2× faster on GPU with competitive accuracy, Native 512×512, smart patching for big images.
efficiency is our product @LiquidAI_
download them on @huggingface:
https://huggingface.co/LiquidAI/LFM2-VL-1.6B
https://huggingface.co/LiquidAI/LFM2-VL-450M
read the blog post: https://www.liquid.ai/blog/lfm2-vl-efficient-vision-language-models
RT @ramin_m_h: meet LFM2-VL: an efficient Liquid vision-language model for the device class. open weights, 440M & 1.6B, up to 2× faster on GPU with competitive accuracy, Native 512×512, smart patching for big images.
efficiency is our product @LiquidAI_
download them on @huggingface:
https://huggingface.co/LiquidAI/LFM2-VL-1.6B
https://huggingface.co/LiquidAI/LFM2-VL-450M
read the blog post: https://www.liquid.ai/blog/lfm2-vl-efficient-vision-language-models
This media is not supported in your browser
VIEW IN TELEGRAM
Hugging Face (Twitter)
RT @jandotai: Introducing Jan-v1: 4B model for web search, an open-source alternative to Perplexity Pro.
In our evals, Jan v1 delivers 91% SimpleQA accuracy, slightly outperforming Perplexity Pro while running fully locally.
Use cases:
- Web search
- Deep Research
Built on the new version of Qwen's Qwen3-4B-Thinking (up to 256k context length), fine-tuned for reasoning and tool use in Jan.
You can run the model in Jan, llama.cpp, or vLLM. To enable search in Jan, go to Settings → Experimental Features → On, then Settings → MCP Servers → enable a search-related MCP such as Serper.
Use the model:
- Jan-v1-4B: https://huggingface.co/janhq/Jan-v1-4B
- Jan-v1-4B-GGUF: https://huggingface.co/janhq/Jan-v1-4B-GGUF
Credit to the @Alibaba_Qwen team for Qwen3 4B Thinking & @ggerganov for llama.cpp.
RT @jandotai: Introducing Jan-v1: 4B model for web search, an open-source alternative to Perplexity Pro.
In our evals, Jan v1 delivers 91% SimpleQA accuracy, slightly outperforming Perplexity Pro while running fully locally.
Use cases:
- Web search
- Deep Research
Built on the new version of Qwen's Qwen3-4B-Thinking (up to 256k context length), fine-tuned for reasoning and tool use in Jan.
You can run the model in Jan, llama.cpp, or vLLM. To enable search in Jan, go to Settings → Experimental Features → On, then Settings → MCP Servers → enable a search-related MCP such as Serper.
Use the model:
- Jan-v1-4B: https://huggingface.co/janhq/Jan-v1-4B
- Jan-v1-4B-GGUF: https://huggingface.co/janhq/Jan-v1-4B-GGUF
Credit to the @Alibaba_Qwen team for Qwen3 4B Thinking & @ggerganov for llama.cpp.
This media is not supported in your browser
VIEW IN TELEGRAM
Hugging Face (Twitter)
RT @Skywork_ai: Matrix-Game 2.0 — The FIRST open-source, real-time, long-sequence interactive world model
Last week, DeepMind's Genie 3 shook the AI world with real-time interactive world models.
But... it wasn't open-sourced.
Today, Matrix-Game 2.0 changed the game. 🚀
25FPS. Minutes-long interaction. Fully open-source.
RT @Skywork_ai: Matrix-Game 2.0 — The FIRST open-source, real-time, long-sequence interactive world model
Last week, DeepMind's Genie 3 shook the AI world with real-time interactive world models.
But... it wasn't open-sourced.
Today, Matrix-Game 2.0 changed the game. 🚀
25FPS. Minutes-long interaction. Fully open-source.
This media is not supported in your browser
VIEW IN TELEGRAM
Hugging Face (Twitter)
RT @reach_vb: Matrix Game 2.0 - Open source, real-time, interactive world model on Hugging Face! 🔥
RT @reach_vb: Matrix Game 2.0 - Open source, real-time, interactive world model on Hugging Face! 🔥
Hugging Face (Twitter)
RT @lhoestq: Let me explain why Hugging Face Datasets storage is faster than S3 + why today's release changes everything 🧵
RT @lhoestq: Let me explain why Hugging Face Datasets storage is faster than S3 + why today's release changes everything 🧵
Hugging Face (Twitter)
RT @kadirnardev: We're releasing a TTS model trained with a 350M parameter and 140,000-hour voice dataset as open source on the Vyvo account tomorrow 🎉 Let's turn on notifications 🔔
RT @kadirnardev: We're releasing a TTS model trained with a 350M parameter and 140,000-hour voice dataset as open source on the Vyvo account tomorrow 🎉 Let's turn on notifications 🔔
Hugging Face (Twitter)
RT @ClementDelangue: Fun to think about open-source models and their variants as families from an evolutionary biology standpoint and analyze "genetic similarity and mutation of traits over model families".
These are the 2,500th, 250th, 50th and 25th largest families on @huggingface: https://twitter.com/didaoh/status/1955381767420121283#m
RT @ClementDelangue: Fun to think about open-source models and their variants as families from an evolutionary biology standpoint and analyze "genetic similarity and mutation of traits over model families".
These are the 2,500th, 250th, 50th and 25th largest families on @huggingface: https://twitter.com/didaoh/status/1955381767420121283#m
This media is not supported in your browser
VIEW IN TELEGRAM
Hugging Face (Twitter)
RT @NVIDIAAIDev: We just released 3 million samples of high quality vision language model training dataset for use cases such as:
📄 optical character recognition (OCR)
📊 visual question answering (VQA)
📝 captioning
🤗 Learn more: nvda.ws/4oyfevu
📥 Download: nvda.ws/4fz2gtB
RT @NVIDIAAIDev: We just released 3 million samples of high quality vision language model training dataset for use cases such as:
📄 optical character recognition (OCR)
📊 visual question answering (VQA)
📝 captioning
🤗 Learn more: nvda.ws/4oyfevu
📥 Download: nvda.ws/4fz2gtB
Hugging Face (Twitter)
RT @jxmnop: OpenAI hasn’t open-sourced a base model since GPT-2 in 2019. they recently released GPT-OSS, which is reasoning-only...
or is it?
turns out that underneath the surface, there is still a strong base model. so we extracted it.
introducing gpt-oss-20b-base 🧵
RT @jxmnop: OpenAI hasn’t open-sourced a base model since GPT-2 in 2019. they recently released GPT-OSS, which is reasoning-only...
or is it?
turns out that underneath the surface, there is still a strong base model. so we extracted it.
introducing gpt-oss-20b-base 🧵
Hugging Face (Twitter)
RT @BrigitteTousi: HAPPENING TODAY: Join @ClementDelangue for an AMA on the Hugging Face Discord!
⏰ 8am PST / 11am EST / 16h CET
🔗 https://discord.com/invite/6r5TEXyk?event=1404451892179763311 https://twitter.com/BrigitteTousi/status/1955300164815462460#m
RT @BrigitteTousi: HAPPENING TODAY: Join @ClementDelangue for an AMA on the Hugging Face Discord!
⏰ 8am PST / 11am EST / 16h CET
🔗 https://discord.com/invite/6r5TEXyk?event=1404451892179763311 https://twitter.com/BrigitteTousi/status/1955300164815462460#m
This media is not supported in your browser
VIEW IN TELEGRAM
Hugging Face (Twitter)
RT @reach_vb: OpenAI gpt-oss 120B orchestrates a full video using Hugging Face spaces! 🤯
All of it, in one SINGLE prompt:
create an image of a Labrador and use it to generate a simple video of it
🛠️ Tools used:
1. Flux.1 Krea Dev by @bfl_ml
2. LTX Fast by @Lightricks
That's it, gpt-oss 120B is one of the BEST open source models I've used for tool calling so far! Kudos @OpenAI 🤗
RT @reach_vb: OpenAI gpt-oss 120B orchestrates a full video using Hugging Face spaces! 🤯
All of it, in one SINGLE prompt:
create an image of a Labrador and use it to generate a simple video of it
🛠️ Tools used:
1. Flux.1 Krea Dev by @bfl_ml
2. LTX Fast by @Lightricks
That's it, gpt-oss 120B is one of the BEST open source models I've used for tool calling so far! Kudos @OpenAI 🤗
Hugging Face (Twitter)
RT @mervenoyann: new TRL comes packed for vision language models 🔥
we shipped support for
> native supervised fine-tuning for VLMs
> multimodal GRPO
> MPO 🫡
read all about it in our blog 🤗 next one!
RT @mervenoyann: new TRL comes packed for vision language models 🔥
we shipped support for
> native supervised fine-tuning for VLMs
> multimodal GRPO
> MPO 🫡
read all about it in our blog 🤗 next one!
Hugging Face (Twitter)
RT @Xianbao_QIAN: A very interesting food dish dataset, if you're building a health app/model: 100 k carefully curated food samples spanning home-cooked meals, restaurant dishes, raw ingredients and packaged products.
How it was built is just as valuable
• 50 k real users on Binance captured their own plates and pre-annotated by professional human annotators.
• Machine-generated labels were then spot-checked and refined by Biance users to guarantee quality.
• A slice of the dataset made available on Hugging Face under an OpenRail license.
Sounds like a new approach for crowdsourcing data collection.
Link below:
RT @Xianbao_QIAN: A very interesting food dish dataset, if you're building a health app/model: 100 k carefully curated food samples spanning home-cooked meals, restaurant dishes, raw ingredients and packaged products.
How it was built is just as valuable
• 50 k real users on Binance captured their own plates and pre-annotated by professional human annotators.
• Machine-generated labels were then spot-checked and refined by Biance users to guarantee quality.
• A slice of the dataset made available on Hugging Face under an OpenRail license.
Sounds like a new approach for crowdsourcing data collection.
Link below:
Hugging Face (Twitter)
RT @allen_ai: With fresh support of $75M from @NSF and $77M from @nvidia, we’re set to scale our open model ecosystem, bolster the infrastructure behind it, and fast‑track reproducible AI research to unlock the next wave of scientific discovery. 💡
RT @allen_ai: With fresh support of $75M from @NSF and $77M from @nvidia, we’re set to scale our open model ecosystem, bolster the infrastructure behind it, and fast‑track reproducible AI research to unlock the next wave of scientific discovery. 💡
Hugging Face (Twitter)
RT @Xianbao_QIAN: A fully open-sourced, top tier Deep Research framework. Guess which one it is?
RT @Xianbao_QIAN: A fully open-sourced, top tier Deep Research framework. Guess which one it is?
Hugging Face (Twitter)
RT @brendanh0gan: introducing qqWen: our fully open-sourced project (code+weights+data+detailed technical report) for full-stack finetuning (pretrain+SFT+RL) a series of models (1.5b, 3b, 7b, 14b & 32b) for a niche financial programming language called Q
All details below!
RT @brendanh0gan: introducing qqWen: our fully open-sourced project (code+weights+data+detailed technical report) for full-stack finetuning (pretrain+SFT+RL) a series of models (1.5b, 3b, 7b, 14b & 32b) for a niche financial programming language called Q
All details below!
This media is not supported in your browser
VIEW IN TELEGRAM
Hugging Face (Twitter)
RT @TencentHunyuan: 🚀We are thrilled to open-source Hunyuan-GameCraft, a high-dynamic interactive game video generation framework built on HunyuanVideo.
It generates playable and physically realistic videos from a single scene image and user action signals, empowering creators and developers to "direct" games with first-person or third-person perspectives.
Key Advantages:
🔹High Dynamics: Unifies standard keyboard inputs into a shared continuous action space, enabling high-precision control over velocity and angle. This allows for the exploration of complex trajectories, overcoming the stiff, limited motion of traditional models. It can also generate dynamic environmental content like moving clouds, rain, snow, and water flow.
🔹Long-term Consistency: Uses hybrid history condition to preserve the original scene information after significant movement.
🔹Significant Cost Reduction: No need for expensive modeling/rendering. PCM distillation compresses...
Перейти на оригинальный пост
RT @TencentHunyuan: 🚀We are thrilled to open-source Hunyuan-GameCraft, a high-dynamic interactive game video generation framework built on HunyuanVideo.
It generates playable and physically realistic videos from a single scene image and user action signals, empowering creators and developers to "direct" games with first-person or third-person perspectives.
Key Advantages:
🔹High Dynamics: Unifies standard keyboard inputs into a shared continuous action space, enabling high-precision control over velocity and angle. This allows for the exploration of complex trajectories, overcoming the stiff, limited motion of traditional models. It can also generate dynamic environmental content like moving clouds, rain, snow, and water flow.
🔹Long-term Consistency: Uses hybrid history condition to preserve the original scene information after significant movement.
🔹Significant Cost Reduction: No need for expensive modeling/rendering. PCM distillation compresses...
Перейти на оригинальный пост