Eagle Corp 开源 EAGLE-4：自适应推测解码加速 LLM 推理
做 LLM 部署的可以看看，EAGLE-4 开源了，推测解码能自适应硬件和负载，Qwen3.8-27B 实测跑到 961.6 tokens/s。
Eagle Corp 发布开源推测解码方案 EAGLE-4，单一 drafter 可根据 batch size、硬件和负载自适应调整解码策略，同时保持目标模型的输出分布。官方使用 Qwen3.8-27B 作为目标模型进行测试，在 4 块 NVIDIA B300 上达到 961.6 tokens/s。Eagle 是一家 AI 推理平台公司，客户包括 NVIDIA、Meta、Intel、AMD 和 Perplexity。
Eagle Corp released EAGLE-4, its latest open-source flexible drafter, for accelerating LLM inference.
> At its core, “One drafter adapts its speculative decoding strategy to the batch size, the hardware, and the workload, and keeps the target model's output distribution.”
> Qwen3.8-27B with its EAGLE-4 draft is the measured target pair, achieving 961.6 tokens/s while running on 4x NVIDIA B300.
Eagle is an AI inference platform that powers leading companies like NVIDIA, Meta, Intel, AMD, and Perplexity.
Inference testing time 👀
