MyColo measured local AI inference

Share card

bundle 33 · measured result · 27B

144.1 tok/s

hardware
NVIDIA GeForce RTX 5090
model
qwen3.8-27b · 22130167c4c20e20…
engine / quant / ctx
ollama · Q4_K_M · ctx 4096
workload
wl-v1-qwen38-27b-ollama
trust
measured · lone · test · 2026-09-21 01:04:34

Throughput comparison

144.1 tok/s · baseline 143.2 · +0.6% · n=10

27B · qwen3.8-27b · your prior signed/measured runs (same recipe)

Same handle, hardware class, size track, and recipe version. Current run excluded.

hash
63b1cdd27ebbeb2fa0df2df679d43ced23d9400d2c70d50edb33fbeb94a23d05

Hardware class

{
  "cpu": "AMD",
  "driver": "580.126.18",
  "gpu_or_soc": "NVIDIA GeForce RTX 5090",
  "os_name": "Linux",
  "stack": "cuda",
  "vram_or_unified_mib": 32607
}

Sign-offs

hardware receipt · JSON · measure