Popular repositories Loading

1. MiroThinker MiroThinker Public

   Forked from MiroMindAI/MiroThinker

   MiroThinker is open-source agentic models trained for deep research and complex tool use scenarios.

   Python

2. unsloth unsloth Public

   Forked from unslothai/unsloth

   Fine-tuning & Reinforcement Learning for LLMs. 🦥 Train OpenAI gpt-oss, Qwen3, Llama 4, DeepSeek-R1, Gemma 3, TTS 2x faster with 70% less VRAM.

   Python