verl/HybridFlow: A Flexible and Efficient RL Post-Training Framework

AI turns documents or topics into real, native PowerPoint decks—with native shapes, transitions and animations, data-backed charts and tables on demand, audio narration from speaker notes, and supp…

Mamba SSM architecture

Twinkle✨: Training workbench to make your model glow.

Mirage Persistent Kernel: Compiling LLMs into a MegaKernel

slime is an LLM post-training framework for RL Scaling.

Puzzles for learning Triton

DLRover: An Automatic Distributed Deep Learning System

Use PEFT or Full-parameter to CPT/SFT/DPO/GRPO 600+ LLMs (Qwen3.6, DeepSeek-V4, GLM-5.1, InternLM3, Llama4, ...) and 300+ MLLMs (Qwen3-VL, Qwen3-Omni, InternVL3.5, Ovis2.5, GLM4.5v, Gemma4, Llava, …