Tencent-Hunyuan-Large is the flagship open-source large language model family from Tencent Hunyuan, offering both pre-trained and instruct (fine-tuned) variants. It is designed with long-context capabilities, quantization support, and high performance on benchmarks across general reasoning, mathematics, language understanding, and Chinese / multilingual tasks. It aims to provide competitive capability with efficient deployment and inference. FP8 quantization support to reduce memory usage (~50%) while maintaining precision. High benchmarking performance on tasks like MMLU, MATH, CMMLU, C-Eval, etc.
Features
- Long context window support: up to 256K tokens in pretrain; 128K tokens for instruct models
- FP8 quantization support to reduce memory usage (~50%) while maintaining precision
- Expert-specific learning rate scaling in training (for mixture or expert architectures)
- High benchmarking performance on tasks like MMLU, MATH, CMMLU, C-Eval, etc.
- Hugging Face format compatibility for fine-tuning / inference using frameworks like hf-deepspeed, plus support for flash attention, efficient operators (TRT-LLM)
- Throughput and efficiency improvements: TRT-LLM backend surpasses vLLM by ~30 %, quantized/inference optimizations included
Project Samples
Follow Tencent-Hunyuan-Large
You Might Also Like
Gen AI apps are built with MongoDB Atlas
MongoDB Atlas is the developer-friendly database used to build, scale, and run gen AI and LLM-powered apps—without needing a separate vector database. Atlas offers built-in vector search, global availability across 115+ regions, and flexible document modeling. Start building AI apps faster, all in one place.
Rate This Project
Login To Rate This Project
User Reviews
Be the first to post a review of Tencent-Hunyuan-Large!