Ornith 1.0 35B

Q4_K_M running on AMD Radeon Pro W5700X ×2 with macOS 26.6.1 Tahoe

PERFORMANCE / TOKENS PER SECONDRDNA 1

Prompt processing703.09tok/s

#80 in index

Token generation49.49tok/s

#78 in index

Run fingerprint

Enough context to make the number useful.

ModelOrnith 1.0 35BMoE architecture

QuantizationQ4_K_MExact submitted variant

GPUAMD Radeon Pro W5700X ×231.97 GB VRAM

GPU architectureRDNA 1Metal backend

Test context512prompt tokens

Model familyMoESigned ToshLLM app · amd-gpu · Metal

ToshLLM versionv0.83.19App 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.