Meta dropped swiss army knives for vision with A2.0 license 👏 > image/video encoders for vision language modelling and spatial understanding (object detection etc) 👏 > The vision LM outperforms InternVL3 and Qwen2.5VL 👏 > They also release gigantic video and image datasets
The authors attempt to come up with single versatile vision encoder to align on diverse set of tasks.
They trained Perception Encoder (PE) Core: a new state-of-the-art family of vision encoders that can be aligned for both vision-language and spatial tasks. For zero-shot image tasks, it outperforms latest sota SigLIP2 👏
> Among fine-tuned ones, first one is PE-Spatial. It's a model to detect bounding boxes, segmentation, depth estimation and it outperforms all other models 😮
> Second one is PLM, Perception Language Model, where they combine PE-Core with Qwen2.5 LM 7B. it outperforms all other models (including InternVL3 which was trained with Qwen2.5LM too!)
The authors release the following checkpoints in sizes base, large and giant:
Authors release following datasets 📑 > PE Video: Gigantic video datasete of 1M videos with 120k expert annotations ⏯️ > PLM-Video and PLM-Image: Human and auto-annotated image and video datasets on region-based tasks > PLM-VideoBench: New video benchmark on MCQA
We’re starting from the foundations of modern generative AI by looking at transformers. This chapter is expanded in depth and features so contains new material like:
FREE and CERTIFIED exam on fundamentals of transformers deeper exploration of transformer architectures and attention mechanisms end -to-end exploration of inference strategies for prefill and decode steps
The course has leveled up in complexity and depth, so this a great time to join in if you want to build you own AI models.
Most of the vision LMs focus on image as a whole, lacking localized references in captions, and not taking in visual prompts (points, boxes, drawings around objects)
DAM addresses this on two levels: new vision backbone that takes in focal crops and the image itself, and a large scale dataset 👀
They generate a dataset by extending existing segmentation and referring expression generation datasets like REFCOCO, by passing in the images and classes to VLMs and generating captions.
Lastly, they also release a new benchmark again with self-supervision, they use an LLM to evaluate the detailed captions focusing on localization 👏
Energy is a massive constraint for AI but do you even know what energy your chatGPT convos are using?
We're trying to change this by releasing ChatUI-energy, the first interface where you see in real-time what energy your AI conversations consume. Great work from @jdelavande powered by spaces & TGI, available for a dozen of open-source models like Llama, Mistral, Qwen, Gemma and more.
New king of open VLMs: InternVL3 takes Qwen 2.5's crown! 👑
InternVL have been a wildly successful series of model : and the latest iteration has just taken back their crown thanks to their superior, natively multimodal vision training pipeline.
➡️ Most of the vision language models (VLMs) these days are built like Frankenstein : take a good text-only Large Language Model (LLM) backbone, stitch a specific vision transformer (ViT) on top of it. Then the training is sequential 🔢 : 1. Freeze the LLM weights while you train the ViT only to work with the LLM part, then 2. Unfreeze all weights to train all weights in order to work together.
💫 The Shanghai Lab decided to challenge this paradigm and chose this approach that they call "native". For each of their model sizes, they still start from a good LLM (mostly Qwen-2.5 series, did I tell you I'm a huge fan of Qwen? ❤️), and stitch the ViT, but they don't freeze anything : they train all weights together with interleaved text and image understanding data in a single pre-training phase 🎨.
They claim it results in more seamless interactions between modalities. And the results prove them right: they took the crown of top VLMs, at nearly all sizes, from their Qwen-2.5 parents. 👑
Hacked my presentation building with inference providers, Cohere command a, and sheer simplicity. Use this script if you’re burning too much time on presentations:
This is what it does: - uses command a to generates slides and speaker notes based on some material. - it renders the material in remark open format and imports all images, tables, etc - you can then review the slides as markdown and iterate - export to either pdf or pptx using backslide
🚀 Next steps are: add text to speech for the audio and generate a video. This should make Hugging Face educational content scale to a billion AI Learners.
You can now bill your inference costs from all our inference partners (together, fireworks, fal, sambanova, cerebras, hyperbolic,...) to your Hugging Face organization.
Useful to drive more company-wide usage of AI without the billing headaches!
Reasoning models like o3 and o4-mini are advancing faster than ever, but imagine what will be possible when they can run locally in your browser! 🤯
Well, with 🤗 Transformers.js, you can do just that! Here's Zyphra's new ZR1 model running at over 100 tokens/second on WebGPU! ⚡️
Giving models access to browser APIs (like File System, Screen Capture, and more) could unlock an entirely new class of web experiences that are personalized, interactive, and run locally in a secure, sandboxed environment.