xtuner
A Next-Generation Training Engine Built for Ultra-Large MoE Models
- 5.2k
- Python
- Apache-2.0
- 14 h ago
A Next-Generation Training Engine Built for Ultra-Large MoE Models
[ICCV 2025] Are VLMs Ready for Autonomous Driving? An Empirical Study from the Reliability, Data, and Metric Perspectives
Higher performance OpenAI LLM service than vLLM serve: A pure C++ high-performance OpenAI LLM service implemented with GPRS+TensorRT-LLM+Tokenizers.cpp, supporting chat and function call, AI agents, distributed multi-GPU inference, multimodal capabilities, and a Gradio chat interface.