Hy MT2 30B

Q6_K running on AMD Radeon Pro 5600M with macOS 15.7.9 Sequoia

PERFORMANCE / TOKENS PER SECONDRDNA 1

Prompt processing99.37tok/s

#187 in index

Token generation8.93tok/s

#188 in index

Run fingerprint

Enough context to make the number useful.

ModelHy MT2 30BMoE architecture

QuantizationQ6_KExact submitted variant

GPUAMD Radeon Pro 5600M7.98 GB VRAM

GPU architectureRDNA 1Metal backend

Test context512prompt tokens

Model familyMoESigned ToshLLM app · ncmoe 43 · amd-gpu · Metal

ToshLLM versionv0.86.3App 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.