Eagle 是一系列前沿视觉语言模型，主打以数据为中心的训练策略，能看图理解并回答多模态问题，适合做图文问答和视觉理解研究。
EAGLE 系列投机解码算法的官方实现，通过草稿模型加速大语言模型推理，在几乎不损失生成质量的前提下显著提升吞吐。
EAGLE-4 的官方推理代码，用投机解码加速大模型生成，可接入 SGLang、vLLM 等主流推理框架，显著提升 token 吞吐。
腾讯开源的模型压缩工具包，支持量化、投机解码和扩散模型加速，帮你在部署大模型时降低显存占用、提升推理效率。
基于 PyTorch 的投机解码模型训练库，支持 EAGLE3、FSDP 等方案，帮你在 vLLM 等推理服务里加速大模型生成。
EAGLE-4 combines a draft model with eagle-harness inside your inference engine to adapt speculative decoding to your hardware, batch and workload. Version 0.1.5 improves
EAGLE-4 combines a draft model with eagle-harness inside your inference engine to adapt speculative decoding to your hardware, batch and workload. Version 0.1.4 makes vLL
EAGLE-4 combines a draft model with eagle-harness inside your inference engine to adapt speculative decoding to your hardware, batch and workload. Version 0.1.3 brings na
EAGLE-4 0.1.2 updates eagle-harness on the 0.1.1 engines. On SGLang, EAGLE-4 decodes a batch without speculation wherever speculation would be slower, so the high-load re
EAGLE-4 0.1.1 updates eagle-harness and two engine builds. An SGLang server now needs only the draft path to serve EAGLE-4 and starts its first requests faster, and a new
…r the univariate drift calculator. Thanks for the eagle-eyed reading [@josecaloca](https://github.com/josecaloca)! [( 425)]((https://github.com/NannyML/nannyml/issues/425