Qwen2.5-Omni is soooo good that people build multimodal reasoning models off of it ๐ฅน > KE-Team/Ke-Omni-R-3B is open-source audio reasoning model sota on average of benchmarks, based on Qwen/Qwen2.5-Omni-3B ๐ฃ๏ธ > Haoz0206/Omni-R1 is a video reasoning model with pixel level grounding (see below) and it's super competitive โฏ๏ธ based on Qwen/Qwen2.5-Omni-7B
vision LMs are saturated over benchmarks, so we built vibe eval ๐ฌ
> compare different models with refreshed in-the-wild examples in different categories ๐ค > submit your favorite model for eval no numbers -- just vibes!
emerging trend: models that can understand image + text and generate image + text
don't miss out โคต๏ธ > MMaDA: single 8B diffusion model aligned with CoT (reasoning!) + UniGRPO Gen-Verse/MMaDA > BAGEL: 7B MoT model based on Qwen2.5, SigLIP-so-400M, Flux VAE ByteDance-Seed/BAGEL both by ByteDance! ๐ฑ
multimodal ๐ฌ๐ผ๏ธ > new moondream (VLM) is out: it's 4-bit quantized (with QAT) version of moondream-2b, runs on 2.5GB VRAM at 184 tps with only 0.6% drop in accuracy (OS) ๐ > ByteDance released BAGEL-7B, an omni model that understands and generates both image + text. they also released Dolphin, a document parsing VLM ๐ฌ (OS) > Google DeepMind dropped MedGemma in I/O, VLM that can interpret medical scans, and Gemma 3n, an omni model with competitive LLM performance
> MMaDa is a new 8B diffusion language model that can generate image and text
LLMs > Mistral released Devstral, a 24B coding assistant (OS) ๐ฉ๐ปโ๐ป > Fairy R1-32B is a new reasoning model -- distilled version of DeepSeek-R1-Distill-Qwen-32B (OS) > NVIDIA released ACEReason-Nemotron-14B, new 14B math and code reasoning model > sarvam-m is a new Indic LM with hybrid thinking mode, based on Mistral Small (OS) > samhitika-0.0.1 is a new Sanskrit corpus (BookCorpus translated with Gemma3-27B)
image generation ๐จ > MTVCrafter is a new human motion animation generator
It's just become easier to share your apps on the biggest AI app store (aka HF spaces) for unlimited storage, more visibility and community interactions.
Just pick a React, Svelte, or Vue template when you create your space or add app_build_command: npm run build in your README's YAML and app_file: build/index.html in your README's YAML block.
This is available today, in the open-source version of phospho. Still is 100% compatible with LeRobot.
The LeRobot dataset by HuggingFace and Remi Cadene is becoming a standard to create robotics datasets. But working with it can rapidly become a nightmare:
- you can't delete a faulty episode. Failed a demo? Finito. - you can't merge datasets - you can't split datasets
So we fixed it.
Now, in the dashboard or in Python, using phospho you can: - repair corrupted LeRobot datasets - delete episodes from a dataset - merge datasets - split datasets
Playing with Veo3 this morning. Share your prompt if you want me to create videos for you (bonus point if they funnily reference HF/open-source). These videos are "a cat on the moon rapping "I love Hugging Face""!
> first reasoning model for robotics > based on Qwen 2.5-VL-7B, use with Hugging Face transformers or vLLM ๐ค > comes with SFT & alignment datasets and a new benchmark ๐
