This media is not supported in your browser
VIEW IN TELEGRAM
Hugging Face (Twitter)
RT @bfl_ml: FLUX.1 Kontext [dev] Hackathon is live!
$10K+ in prizes, open worldwide. 7 days to experiment and surprise us. Create LoRAs, build workflows, or try something totally unexpected.
Run it locally or through our partners @NVIDIA_AI_PC @fal @huggingface
Registration link below π
RT @bfl_ml: FLUX.1 Kontext [dev] Hackathon is live!
$10K+ in prizes, open worldwide. 7 days to experiment and surprise us. Create LoRAs, build workflows, or try something totally unexpected.
Run it locally or through our partners @NVIDIA_AI_PC @fal @huggingface
Registration link below π
This media is not supported in your browser
VIEW IN TELEGRAM
Hugging Face (Twitter)
RT @DecartAI: We are building βOpen Source Nano Banana for Videoβ - here is open source demo v0.1
We are open sourcing Lucy Edit, the first foundation model for text-guided video editing!
Get the model on @huggingface π€, API on @FAL, and nodes on @ComfyUI π§΅
RT @DecartAI: We are building βOpen Source Nano Banana for Videoβ - here is open source demo v0.1
We are open sourcing Lucy Edit, the first foundation model for text-guided video editing!
Get the model on @huggingface π€, API on @FAL, and nodes on @ComfyUI π§΅
This media is not supported in your browser
VIEW IN TELEGRAM
Hugging Face (Twitter)
RT @Xianbao_QIAN: WAN 2.2 animate model & demo is now officially released on @huggingface
RT @Xianbao_QIAN: WAN 2.2 animate model & demo is now officially released on @huggingface
βHugging Face (Twitter)
RT @ariG23498: The new kid in the blog for experiment tracking is trackio.
And here you have @abidlabs talk about it.
https://www.youtube.com/watch?v=BdS8FgBqNOM?si=bbrQ89X7677rontC
RT @ariG23498: The new kid in the blog for experiment tracking is trackio.
And here you have @abidlabs talk about it.
https://www.youtube.com/watch?v=BdS8FgBqNOM?si=bbrQ89X7677rontC
YouTube
Trackio: A DROP-IN Replacement for W&B that is open-source and π― free
This video provides an overview and demo of Trackio, which is a free, experiment tracking library Hugging Face just released.
Install Trackio: pip install trackio
Documentation: https://huggingface.co/docs/trackio/index
Install Trackio: pip install trackio
Documentation: https://huggingface.co/docs/trackio/index
Media is too big
VIEW IN TELEGRAM
Hugging Face (Twitter)
RT @abidlabs: BOOM! A new, free experiment tracking library with identical syntax as wandb that makes it trivial as a drop-in replacement
RT @abidlabs: BOOM! A new, free experiment tracking library with identical syntax as wandb that makes it trivial as a drop-in replacement
Hugging Face (Twitter)
RT @_akhaliq: moondream3-preview is out on Hugging Face
vision language model with a mixture-of-experts architecture (9B total parameters, 2B active)
delivering sota visual reasoning while still being efficient and deployment-friendly
vibe coded a quick app for it in anycoder
RT @_akhaliq: moondream3-preview is out on Hugging Face
vision language model with a mixture-of-experts architecture (9B total parameters, 2B active)
delivering sota visual reasoning while still being efficient and deployment-friendly
vibe coded a quick app for it in anycoder
βHugging Face (Twitter)
RT @AdinaYakup: MiMo-Audio π Open audio model released by @Xiaomi
https://huggingface.co/collections/XiaomiMiMo/mimo-audio-68cc7202692c27dae881cce0
β¨ 7B base & instruct - MIT license
β¨ Pretrained on 100M+ hours
β¨ Few-shot across speech & audio tasks
RT @AdinaYakup: MiMo-Audio π Open audio model released by @Xiaomi
https://huggingface.co/collections/XiaomiMiMo/mimo-audio-68cc7202692c27dae881cce0
β¨ 7B base & instruct - MIT license
β¨ Pretrained on 100M+ hours
β¨ Few-shot across speech & audio tasks
huggingface.co
MiMo-Audio - a XiaomiMiMo Collection
Weβre on a journey to advance and democratize artificial intelligence through open source and open science.
Hugging Face (Twitter)
RT @adibvafa: CodonTransformer, our open-soruce model on @huggingface that optimizes genes for protein expression has passed 250,000+ downloads!
RT @adibvafa: CodonTransformer, our open-soruce model on @huggingface that optimizes genes for protein expression has passed 250,000+ downloads!
Hugging Face (Twitter)
RT @XiaomiMiMo: π Say Hi to MiMo-Audio!
Our BREAKTHROUGH in general-purpose audio intelligence.
π― Scaling pretraining to 100M+ hours leads to EMERGENCE of few-shot generalization across diverse audio tasks!
π₯ Post-trained MiMo-Audio-7B-Instruct:
β’ crushes benchmarks: SOTA on MMSU, MMAU, MMAR, MMAU-Pro
β’ outperforms Gemini-2.5-Flash on audio understanding
β’ beats GPT-4o-Audio on complex reasoning tasks
π The best part? It's 100% OPEN-SOURCE
Everything from tokenizer to model to evaluations!
π€ Try it in HF Space: https://huggingface.co/spaces/XiaomiMiMo/mimo_audio_chat
π Tech Blog: https://xiaomimimo.github.io/MiMo-Audio-Demo/
RT @XiaomiMiMo: π Say Hi to MiMo-Audio!
Our BREAKTHROUGH in general-purpose audio intelligence.
π― Scaling pretraining to 100M+ hours leads to EMERGENCE of few-shot generalization across diverse audio tasks!
π₯ Post-trained MiMo-Audio-7B-Instruct:
β’ crushes benchmarks: SOTA on MMSU, MMAU, MMAR, MMAU-Pro
β’ outperforms Gemini-2.5-Flash on audio understanding
β’ beats GPT-4o-Audio on complex reasoning tasks
π The best part? It's 100% OPEN-SOURCE
Everything from tokenizer to model to evaluations!
π€ Try it in HF Space: https://huggingface.co/spaces/XiaomiMiMo/mimo_audio_chat
π Tech Blog: https://xiaomimimo.github.io/MiMo-Audio-Demo/
This media is not supported in your browser
VIEW IN TELEGRAM
Hugging Face (Twitter)
RT @LeRobotHF: LeRobot SO101 setup just got 50% cheaper!
You can now teleoperate your follower arm right from your phone. π€―
But that's not all. Our new pipeline feature lets you record and train AI models in end-effector space, or with any other features. The possibilities are endless!
RT @LeRobotHF: LeRobot SO101 setup just got 50% cheaper!
You can now teleoperate your follower arm right from your phone. π€―
But that's not all. Our new pipeline feature lets you record and train AI models in end-effector space, or with any other features. The possibilities are endless!
Hugging Face (Twitter)
RT @Baidu_Inc: Qianfan-VL, Baidu AI Cloud's vision-language model series, is now open source! Designed for enterprise-level applications, these multimodal models combine robust general capabilities with advanced performance in OCR and math problem-solving.
Key features:
> Three model sizes (3B, 8B, 70B) with 32K context length for diverse needs
> Chain-of-thought reasoning in 8B/70B for strong performance in chart understanding, math, and visual logic
> Four-stage progressive training pipeline for improved cross-modal alignment and domain enhancement
> High-precision data synthesis pipeline across documents, math, charts, tables, formulas, and OCR tasks
Discover more about Qianfan-VL β
RT @Baidu_Inc: Qianfan-VL, Baidu AI Cloud's vision-language model series, is now open source! Designed for enterprise-level applications, these multimodal models combine robust general capabilities with advanced performance in OCR and math problem-solving.
Key features:
> Three model sizes (3B, 8B, 70B) with 32K context length for diverse needs
> Chain-of-thought reasoning in 8B/70B for strong performance in chart understanding, math, and visual logic
> Four-stage progressive training pipeline for improved cross-modal alignment and domain enhancement
> High-precision data synthesis pipeline across documents, math, charts, tables, formulas, and OCR tasks
Discover more about Qianfan-VL β
Hugging Face (Twitter)
RT @mervenoyann: this summer we have shipped a ton of things in TRL! π₯ποΈπ
try out bleeding-edge fine-tuning methods with few lines of CLI commands and check out notebooks to get started π€
RT @mervenoyann: this summer we have shipped a ton of things in TRL! π₯ποΈπ
try out bleeding-edge fine-tuning methods with few lines of CLI commands and check out notebooks to get started π€
Hugging Face (Twitter)
RT @abidlabs: I'm interested in hiring a python engineer who knows @Gradio well & likes experimenting with many different projects simultaneously and growing the ones that are the most impactful.
DM if you'd like to work with me @huggingface, and share your most impressive Gradio app.
RT @abidlabs: I'm interested in hiring a python engineer who knows @Gradio well & likes experimenting with many different projects simultaneously and growing the ones that are the most impactful.
DM if you'd like to work with me @huggingface, and share your most impressive Gradio app.
Hugging Face (Twitter)
RT @ClementDelangue: Xet by Hugging Face is the most important AI technology that nobody is talking about!
Under the hood, it now powers 5M Xet-enabled AI models & datasets on HF which see hundreds of terabytes of uploads and downloads every single day.
What makes it super powerful is that it massively speeds up & reduces costs of data transfer thanks to methods like content-defined chunking (CDC). Instead of treating a file as an indivisible unit, CDC breaks files down into variable-sized chunks, using the data to define boundaries.
That's what allows @huggingface to offer a platform for 10 million AI builders in open-source at a fraction of the cost.
Thanks @xetdata team!
RT @ClementDelangue: Xet by Hugging Face is the most important AI technology that nobody is talking about!
Under the hood, it now powers 5M Xet-enabled AI models & datasets on HF which see hundreds of terabytes of uploads and downloads every single day.
What makes it super powerful is that it massively speeds up & reduces costs of data transfer thanks to methods like content-defined chunking (CDC). Instead of treating a file as an indivisible unit, CDC breaks files down into variable-sized chunks, using the data to define boundaries.
That's what allows @huggingface to offer a platform for 10 million AI builders in open-source at a fraction of the cost.
Thanks @xetdata team!
Hugging Face (Twitter)
RT @ClementDelangue: Granite Docling by @IBM is #3 trending on @huggingface.
This is a multimodal Image-Text-to-Text model engineered for efficient document conversion. It preserves the core features of Docling while maintaining seamless integration with DoclingDocuments to ensure full compatibility.
It builds upon the IDEFICS3 architecture, but introduces two key modifications: it replaces the vision encoder with siglip2-base-patch16-512 and substitutes the language model with a Granite 165M LLM. Try out our Granite-Docling-258 demo today.
License: Apache 2.0
Granite-docling-258M is fully integrated into the Docling pipelines, carrying over existing features while introducing a number of powerful new features, including:
π’ Enhanced Equation Recognition: More accurate detection and formatting of mathematical formulas
𧩠Flexible Inference Modes: Choose between full-page inference, bbox-guided region inference
π§ Improved Stability: Tends to avoid...
ΠΠ΅ΡΠ΅ΠΉΡΠΈ Π½Π° ΠΎΡΠΈΠ³ΠΈΠ½Π°Π»ΡΠ½ΡΠΉ ΠΏΠΎΡΡ
RT @ClementDelangue: Granite Docling by @IBM is #3 trending on @huggingface.
This is a multimodal Image-Text-to-Text model engineered for efficient document conversion. It preserves the core features of Docling while maintaining seamless integration with DoclingDocuments to ensure full compatibility.
It builds upon the IDEFICS3 architecture, but introduces two key modifications: it replaces the vision encoder with siglip2-base-patch16-512 and substitutes the language model with a Granite 165M LLM. Try out our Granite-Docling-258 demo today.
License: Apache 2.0
Granite-docling-258M is fully integrated into the Docling pipelines, carrying over existing features while introducing a number of powerful new features, including:
π’ Enhanced Equation Recognition: More accurate detection and formatting of mathematical formulas
𧩠Flexible Inference Modes: Choose between full-page inference, bbox-guided region inference
π§ Improved Stability: Tends to avoid...
ΠΠ΅ΡΠ΅ΠΉΡΠΈ Π½Π° ΠΎΡΠΈΠ³ΠΈΠ½Π°Π»ΡΠ½ΡΠΉ ΠΏΠΎΡΡ
Hugging Face (Twitter)
RT @Alibaba_Qwen: π Introducing Qwen3-Omni β the first natively end-to-end omni-modal AI unifying text, image, audio & video in one model β no modality trade-offs!
π SOTA on 22/36 audio & AV benchmarks
π 119L text / 19L speech in / 10L speech out
β‘ 211ms latency | π§ 30-min audio understanding
π¨ Fully customizable via system prompts
π Built-in tool calling
π€ Open-source Captioner model (low-hallucination!)
π Whatβs Open-Sourced?
Weβve open-sourced Qwen3-Omni-30B-A3B-Instruct, Qwen3-Omni-30B-A3B-Thinking, and Qwen3-Omni-30B-A3B-Captioner, to empower developers to explore a variety of applications from instruction-following to creative tasks.
Try it now π
π¬ Qwen Chat: https://chat.qwen.ai/?models=qwen3-omni-flash
π» GitHub: github.com/QwenLM/Qwen3-Omni
π€ HF Models: https://huggingface.co/collections/Qwen/qwen3-omni-68d100a86cd0906843ceccbe
π€ MS Models:
https://modelscope.cn/collections/Qwen3-Omni-867aef131e7d4f
π¬ Demo: https://huggingface.co/spaces/Qwen/Qwen3-Omni-Demo
RT @Alibaba_Qwen: π Introducing Qwen3-Omni β the first natively end-to-end omni-modal AI unifying text, image, audio & video in one model β no modality trade-offs!
π SOTA on 22/36 audio & AV benchmarks
π 119L text / 19L speech in / 10L speech out
β‘ 211ms latency | π§ 30-min audio understanding
π¨ Fully customizable via system prompts
π Built-in tool calling
π€ Open-source Captioner model (low-hallucination!)
π Whatβs Open-Sourced?
Weβve open-sourced Qwen3-Omni-30B-A3B-Instruct, Qwen3-Omni-30B-A3B-Thinking, and Qwen3-Omni-30B-A3B-Captioner, to empower developers to explore a variety of applications from instruction-following to creative tasks.
Try it now π
π¬ Qwen Chat: https://chat.qwen.ai/?models=qwen3-omni-flash
π» GitHub: github.com/QwenLM/Qwen3-Omni
π€ HF Models: https://huggingface.co/collections/Qwen/qwen3-omni-68d100a86cd0906843ceccbe
π€ MS Models:
https://modelscope.cn/collections/Qwen3-Omni-867aef131e7d4f
π¬ Demo: https://huggingface.co/spaces/Qwen/Qwen3-Omni-Demo
βHugging Face (Twitter)
RT @AdinaYakup: 3 releases in one day π€― just before Alibaba Cloudβs annual conference! @Alibaba_Qwen is on fire π₯
huggingface.co/Qwen
β¨ Qwen3 Omni: End-to-end omni model
β¨ Qwen3 TTS: Supports CN/EN/IT/FR + 10 langs
β¨ Qwen-Image-Edit-2509: Big upgrade from the previous version
Excited to see whatβs coming in the next 3 days π
RT @AdinaYakup: 3 releases in one day π€― just before Alibaba Cloudβs annual conference! @Alibaba_Qwen is on fire π₯
huggingface.co/Qwen
β¨ Qwen3 Omni: End-to-end omni model
β¨ Qwen3 TTS: Supports CN/EN/IT/FR + 10 langs
β¨ Qwen-Image-Edit-2509: Big upgrade from the previous version
Excited to see whatβs coming in the next 3 days π
huggingface.co
Qwen (Qwen)
Org profile for Qwen on Hugging Face, the AI community building the future.
This media is not supported in your browser
VIEW IN TELEGRAM
Hugging Face (Twitter)
RT @amir_mahla: LETβS GOOO π₯ Just released Smol2Operator, a full open-source recipe for turning a 2.2B model into an agentic GUI coder, and all the tools you need to build your own π«‘
RT @amir_mahla: LETβS GOOO π₯ Just released Smol2Operator, a full open-source recipe for turning a 2.2B model into an agentic GUI coder, and all the tools you need to build your own π«‘