Qwen3.6 27B
on NVIDIA RTX 5090 · vLLM · 262,144 ctx
May 1, 2026
Summary
User reports Qwen3.6-27B-INT4 at 105-108 t/s generation on an RTX 5090 at 256k context.
Setup is vLLM 0.19 with the Lorbus AutoRound quant, an fp8_e4m3 KV cache, flashinfer attention, and MTP speculative decoding with 3 tokens.