Running on CPU Upgrade Featured 3.23k The Smol Training Playbook π 3.23k The secrets to building world-class LLMs
deepseek-ai/DeepSeek-V3.1-Terminus Text Generation β’ 685B β’ Updated Sep 29, 2025 β’ 10.5k β’ β’ 366
Alibaba-NLP/Tongyi-DeepResearch-30B-A3B Text Generation β’ 31B β’ Updated Oct 10, 2025 β’ 63k β’ 812
nvidia/Llama-3_3-Nemotron-Super-49B-v1_5 Text Generation β’ 50B β’ Updated Oct 15, 2025 β’ 54.1k β’ β’ 234
Running Agents 68 KVPress Leaderboard π₯ 68 KVPress leaderboard: benchmark KV Cache compression methods