GPT-1900 Collection Pre-1900 LLMs for physics reasoning. RL models are physics-only; use the SFT model for general chat. Tune temperature (0.6-0.7). • 11 items • Updated Apr 2 • 9
GameWorld: Towards Standardized and Verifiable Evaluation of Multimodal Game Agents Paper • 2604.07429 • Published 28 days ago • 119
FIPO: Eliciting Deep Reasoning with Future-KL Influenced Policy Optimization Paper • 2603.19835 • Published Mar 20 • 350
Running on CPU Upgrade 231 The Synthetic Data Playbook: Generating Trillions of the Finest Tokens 📝 231 Explore synthetic data experiments on a virtual bookshelf
view post Post 2301 We just released our latest Shisa V2.1 Japanese multi-lingual models: https://huggingface.co/collections/shisa-ai/shisa-v21Besides updates to our 14B, and 70B, we have a new LFM2-based 1.2B, Llama 3.2-based 3B, and Qwen 3-based 8B, all with class-leading Japanese language capabilities.Per usual, lots of details in the Model Cards for those interested. See translation 1 reply · 🔥 5 5 + Reply
Running on CPU Upgrade Featured 3.15k The Smol Training Playbook 📚 3.15k The secrets to building world-class LLMs
Less is More: Recursive Reasoning with Tiny Networks Paper • 2510.04871 • Published Oct 6, 2025 • 514