Introduce We trained a qwen2.5 vl 7b eagle3 draft model on 30k data random select from FreedomIntelligence/ALLaVA 4V with specforge Usage infer with sglang benchmark with mmstar start server: benchmark: Speedup with eagle without eagle e2e speed up 1.5x train follow the train qwen2.5 vl eagle3
We use cookies for essential functionality and analytics. You can accept or reject analytics cookies.Cookie policy