EAGLE-4 speculative decoding hits 961.6 tokens/s on Qwen3.8-27B, code open-sourced
zhyncs42 · x · 2026-10-21
Eagle AI Corp released EAGLE-4, the newest generation of its speculative decoding family, pitched as "one flexible drafter for all batch sizes and hardware." It reaches 961.6 tokens/s on Qwen3.8-27B (BF16, 4×B300, TP4, single request, T=0) with TokenSpeed. Official inference code, the paper PDF and Qwen3.8-27B drafter weights are on GitHub, targeting the vLLM/SGLang ecosystem.
Related event: EAGLE-4 Open-Source Speculative Decoding Hits 961.6 Tokens/s(2 posts)→
More from Infra
- ai& claims to be Japan's biggest dedicated inference platform with 100MW+ data centers — DavidBennett__ · 2026-10-21
- SpaceX Reportedly Adding 10GW of Compute by 2027, Up from 1.7GW Today — beffjezos · 2026-10-21
- Musk's Hunt for Rentable GPU Compute Turns Into a Real-Life Cheese Shop Sketch — XFreeze · 2026-10-21
- Cloudflare discloses compromised eMudhra intermediate certificates, urges patching — Cloudflare Blog · 2026-10-21
- Shotwell: Starlink now contracted for ~12,000 planes, a quarter of available aircraft — XFreeze · 2026-10-21
- Samsung Electro-Mechanics signs $260M substrate deal with Micron as AI shortage spreads down the stack — demian_ai · 2026-10-21