DeepSeek-V4.1-Flash: Pushing the Limits of KV Cache Compression Paper • 2609.19969 • Published Sep 17 • 230
view article Article Continuous batching from first principles +1 ror, ArthurZ, mcpotato • Nov 25, 2025 • 455
Running 4.09k The Ultra-Scale Playbook 🌌 4.09k The ultimate guide to training LLM on large GPU Clusters