Files
amd-strix-halo-toolboxes/benchmark/results/29-03-2026/Qwen3-Coder-30B-A3B-Instruct-Q4_K_M__rocm-7_2__fa1.log
T
Donato Capitella 7aa6e6dea9 update benchmarks
2026-04-11 11:18:45 +01:00

9 lines
716 B
Plaintext

ggml_cuda_init: found 1 ROCm devices (Total VRAM: 126976 MiB):
Device 0: Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32, VRAM: 126976 MiB
| model | size | params | backend | ngl | fa | mmap | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
| qwen3moe 30B.A3B Q4_K - Medium | 17.35 GiB | 30.53 B | ROCm | 99 | 1 | 0 | pp512 | 1149.31 ± 13.28 |
| qwen3moe 30B.A3B Q4_K - Medium | 17.35 GiB | 30.53 B | ROCm | 99 | 1 | 0 | tg128 | 67.72 ± 0.02 |
build: 2405d59cb (8577)