Granite 4.1 LLMs: How They Are Built
Granite 4.1 are compact language models from IBM with 3B, 30B and 83B parameters, trained on 15 trillion tokens with a 512K context window. The 8B Instruct model outperforms the larger predecessor model through optimized dense architecture and advanced fine-tuning and reinforcement learning techniques.
Adding Benchmaxxer Repellant to the Open ASR Leaderboard
This post reports on anti-benchmaxxer measures introduced to the open ASR leaderboard, featuring updates from contributors including Eric Bezzam, Steven Zheng, Eustache Le Bihan, Sergio Bruccoleri, and Jeanine Sinanan-Singh from the Appen AI Research team, linked to the Hugging Face platform.
Building Blocks for Foundation Model Training and Inference on AWS
Foundation model development today scales across three channels: pre-training, post-training, and test-time compute, with AWS showing how its infrastructure—accelerators, networking, and storage—works with open-source tools like PyTorch, Kubernetes, and Prometheus to enable efficient training and inference.
