Anybody can now post-train Llama 3.2 Vision on their own dataset in just a few lines of code with TRL 🚀! We've just added support for the 11B and 90B models to the SFTTrainer, so you can fine-tune your models to both see 👀 and follow your instructions ✍️ Training script

1,000,000 model repos! It's time to build, accelerate and repeat! ⚡ open/acc

Run Llama 3.2 1B & 3B in a FREE Google Colab! 🔥 Powered by Transformers ⚡

Let's bring llama.cpp to the clouds! You can now run llama.cpp-powered inference endpoints through Hugging Face with just a few clicks. Simply select a GGUF model, pick your cloud provider (AWS, Azure, GCP), a suitable node GPU/CPU and you are good to go. For more info, check

🚨 Nvidia NeMo 10x faster & 4.5x more cost effective and blows OpenAI Whisper out of the water! 🔥 Tops the Open ASR Leaderboard with a series of CTC, RNN-T, TDT and AED models! What powers the upto 10x speed boost: > Autocasting weights to bfloat16 > CUDA Graphs w/ conditional

“The root of great work is a sort of ambitious curiosity, and you can't manufacture that.”

Open Source AI was off the charts last week: Nvidia released Nemotron 51B - 220% faster and can handle 400% more workload than L3.1 70B & permissively licensed Meta dropped Llama 3.2 - Llama Vision 90B & 11B and tiny llamas (3B & 1B) for on-device usage, multilingual & with