Hf 26B

Q4_0 running on AMD Radeon RX Vega 56 + AMD Radeon Pro WX 5100 with macOS 15.7.9 Sequoia

PERFORMANCE / TOKENS PER SECONDGCN / Vega

Prompt processing111.45tok/s

#163 in index

Token generation20.09tok/s

#135 in index

Run fingerprint

Enough context to make the number useful.

ModelHf 26BMoE architecture

QuantizationQ4_0Exact submitted variant

GPUAMD Radeon RX Vega 56 + AMD Radeon Pro WX 510015.98 GB VRAM

GPU architectureGCN / VegaMetal backend

Test context512prompt tokens

Model familyMoESigned ToshLLM app · ncmoe 2 · amd-gpu · Metal

ToshLLM versionv0.85.8App version used to capture and sign this run

Prompt processing and generation are reported separately so memory bandwidth and decode behavior stay visible.

Quantization, GPU family, VRAM and context travel with the result instead of being hidden behind one score.

This result remains explicitly labeled as a community submission.