Skip to content
llm-speed

Flint-40B-A2.5B-UD-cpumoe12-ctx8k on RTX A4000 (16GB) + AMD EPYC 7453 28-Core Processor (56c) + 504GB

MeasuredOn 2026-10-22, Flint-40B-A2.5B-UD-cpumoe12-ctx… Q3_K_XL ran at 15.7 tok/s (chat-short) on an RTX A4000 (16GB) with llama.cpp. Run r_acy-a8qic5j.

Workload results

WorkloadBackendModeldecode tok/sprefill tok/sTTFTp50p95
chat-shortllama.cpp@0.7.0-dev (build 11803, commit a37cc4610)Flint-40B-A2.5B-UD-cpumoe12-ctx8kQ3_K_XL15.67tok/s190.0tok/s516ms85.7ms99.8ms
chat-longllama.cpp@0.7.0-dev (build 11803, commit a37cc4610)Flint-40B-A2.5B-UD-cpumoe12-ctx8kQ3_K_XL10.01tok/s840.9tok/s3,584ms101ms174ms
concurrent-decodellama.cpp@0.7.0-dev (build 11803, commit a37cc4610)Flint-40B-A2.5B-UD-cpumoe12-ctx8kQ3_K_XL10.08tok/s——102ms106ms
agent-tracellama.cpp@0.7.0-dev (build 11803, commit a37cc4610)Flint-40B-A2.5B-UD-cpumoe12-ctx8kQ3_K_XL10.04tok/s1,988.6tok/s1,028ms100ms174ms

Reproduce on your machine

Start with this model and workload selection. This command does not capture the original artifact, backend version or server settings:

Match quantization, context capacity, caching and runtime settings for a useful comparison. Duration varies by workload and hardware; the published client may differ from this run's version. How it's measured.

Share this run

Copy the short link, a Reddit-ready summary, or a README badge. Every number is the measured result on this page, and each badge links back to it.

llm-speed: 15.7 tok/s on RTX A4000 (16GB) (Flint-40B-A2.5B-UD-cpumoe12-ctx…)
https://llm-speed.com/r/r_acy-a8qic5j

Related benchmarks

Provenance

Run ID
r_acy-a8qic5j
Fingerprint hash
Public key
IS7fEFWW2WxPAFmM9EagBBDr+LYlpSMSYtrEPP/my+E=
Received
2026-10-22 01:32:00