qwen3.6

on RTX 4090

44.2

tok/sdecode

RTX 4090 (48GB) + AMD EPYC 7763 64-Core Processor (128c) + 1008GB

rank in tier

28/31RTX 4090 runstop 90%

best workload

agent-trace

where the rig flew

slowest workload

—

single-workload run

backend

ollamaollama is llama.cpp under the hood; pure llama.cpp tends to nudge a few % faster.

faster than

Qwen3.6-27B on RTX 309042.7 tok/s
Qwen3-32B on RTX 309038.5 tok/s
Qwen3.8-27B on RTX 309037.6 tok/s

This is a shareable view of one signed llm-speed suite-suite-v1 run. Numbers are not edited. The raw record has every workload, the public key, and the fingerprint hash.