llama3.1

on RTX 4090

137.3

tok/sdecode

RTX 4090 (24GB) + AMD EPYC 7642 48-Core Processor (48c) + 252GB

rank in tier

11/31RTX 4090 runstop 35%

best workload

chat-short

where the rig flew

slowest workload

chat-long129.2 tok/s

where the rig struggled

backend

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

faster than

glm-4.7-flash on RTX 3090134.6 tok/s
Qwen2.5-14B-Instruct on RTX 5090133.3 tok/s
Llama-3.1-8B-Instruct-4bit on M3 Ultra130.2 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.