Best use case for local vision agents?
What's the most influential model since BERT?
Nivida just open sourced their long context goodies - 128k context for 50% less memory
FREE Open AI o3 like reasoning model on Hugging Chat - Deepseek R1
Hugging Face will teach you how to use Langchain for agents
Hugging Face Spaces make the perfect agent tools!
How would you build an LLM agent application without using LangChain?
Hugging Face is doing a FREE and CERTIFIED course on LLM Agents!
Qwen released a 72B and a 7B process reward models (PRM) on their recent math models
Hugging Face released a free course on agents.
Kyutai drops Helium 2B Preview - Multilingual Base LLM - CC-BY license 🔥
You don't need agents for code AI, according to Agentless. Is anybody using this approach?
Text to speech in 82m params is perfect for edge AI. Who's building an audio assistant with Kokoro?
WebGPU-accelerated reasoning LLMs running 100% locally in-browser w/ Transformers.js
Why is nobody going wild for bytedance's vision language + segmentation model?
Is di2 worth it for a winter training bike?
[D] Graduated from MIT with a PhD | Teaching you how to build an entire LLM from scratch
Ai Roleplay Characters
Energy efficiency of 5090 is slightly worse than 4090
Former OpenAI employee Miles Brundage: "o1 is just an LLM though, no reasoning infrastructure. The reasoning is in the chain of thought." Current OpenAI employee roon: "Miles literally knows what o1 does."
Anyone want the script to run Moondream 2b's new gaze detection on any video?
Phi 4 is just 14B But Better than llama 3.1 70b for several tasks.
Why is nobody going crazy for this grounded VLM?
New Moondream 2B vision language model release
