Agaram 2.5B: The First Tamil-First Language Model Trained from Scratch with Tool Calling. A Playbook for Under-Represented Languages Paper โข 2610.18420 โข Published Oct 13 โข 7
Running on CPU Upgrade Featured 3.4k The Smol Training Playbook ๐ 3.4k The secrets to building world-class LLMs
SFT Memorizes, RL Generalizes: A Comparative Study of Foundation Model Post-training Paper โข 2501.17161 โข Published Jan 28, 2025 โข 123