Llama 3.3 70B

UD-IQ1_M running on AMD Radeon PRO W6800X Duo ×2 with macOS 26.6.1 Tahoe

PERFORMANCE / TOKENS PER SECONDRDNA 2

Prompt processing291.49tok/s

#130 in index

Token generation8.6tok/s

#194 in index

Run fingerprint

Enough context to make the number useful.

ModelLlama 3.3 70BDense architecture

QuantizationUD-IQ1_MExact submitted variant

GPUAMD Radeon PRO W6800X Duo ×2127.94 GB VRAM

GPU architectureRDNA 2Metal backend

Test context512prompt tokens

Model familyDenseSigned ToshLLM app · amd-gpu · Metal

ToshLLM versionv0.85.7App 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.