Event ·Multi-source

Eagle Corp 开源 EAGLE-4：自适应推测解码加速 LLM 推理

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。

Current read

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。

11 sources58° AI heatLast updated 2026/10/20 22:02:03

Evidence

11 sources

Primary sourceTestingCatalog

2026/10/20 22:02:03View source

Related source 1berryxia

2026/10/20 11:34:11View source

Related source 2Artificial Analysis

2026/10/20 15:20:00View source

Related source 3techcrunch

2026/10/19 15:37:09View source

Related source 4PyTorch

2026/10/20 00:43:25View source

Related source 5marktechpost

2026/10/20 14:31:59View source

Related source 6Yuchen Jin

2026/10/20 17:05:36View source

Related source 7Nathan Lambert

2026/10/20 17:11:08View source

Related source 8SuperTechFans

2026/10/21 00:16:33View source

Related source 9IT之家

2026/10/21 10:39:12View source

Related source 10AI21 Labs

2026/10/21 11:30:03View source

Conflict check

Dedup data only shows that these sources discuss the same event, not that they agree. There is no structured supporting or refuting evidence yet, so no conflict is inferred automatically.

Open the story