EAGLE-4

AISantiago reposts EagleCorp's announcement of EAGLE-4, which it calls its fastest version yet. EagleCorp reports 961.6 tokens/s on Qwen3.8-27B using TokenSpeed with BF16, four NVIDIA B300 GPUs, TP4, one request, and T=0. The post explains that a smaller helper model guesses several upcoming tokens and the main model checks them together, and says the code is available to run with vLLM and SGLang.

Why it matters: Adapts its speculative decoding to batch size, hardware and workload, per Eagle Corp

Descriptions are prepared from the linked sources. Inclusion is not an endorsement.