fal· @fal · X·· 4 小时前AI 评分54
AI 导读
fal 宣布与 Fireworks AI 合作提前上线 EAGLE-4，并发布技术博客说明其工作原理及对创作工作流的意义。引用内容显示，EAGLE-4 在 Qwen3.8-27B 上配合 TokenSpeed 达到 961.6 tokens/s，配置为 BF16、4× NVIDIA B300、TP4、单请求、T=0，代码已在 GitHub 开源。
正文
We are excited to work together with @FireworksAI_HQ on the early-launch of EAGLE-4.
Check out our technical blog to understand how EAGLE-4 works and why it matters for creative workflows: https://blog.fal.ai/eagle-4-technical-deepdive/
EAGLE-4 is here. From EAGLE-1 to EAGLE-3, we’ve kept making inference faster. EAGLE-4 is our fastest yet. 961.6 tokens/s on @Alibaba_Qwen’s Qwen3.8-27B with TokenSpeed (BF16, 4× @NVIDIA B300, TP4, one request, T=0). Github: https://github.com/eagle-hq/EAGLE-4在 X 查看被引用的帖子
来源：fal · x.com