Qwen3-32B-Instruct.Q4_K_M

on smoke-host

10.0

tok/sdecode

smoke-host

rank in tier

1/1smoke-host runs

best workload

chat-short

where the rig flew

slowest workload

—

single-workload run

backend

llama.cppllama.cpp is the universal backend — broad model support, modest speed ceiling.

faster than

Qwen3-32B-4bit on M3 Pro7.2 tok/s
Qwen2.5-32B-Instruct-4bit on M3 Pro7.1 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.