Files
amd-strix-halo-toolboxes/benchmark/results/llama-2-7b.Q4_0__rocm-7_2_3__fa1__longctx32768.log
T
2026-05-12 12:32:07 +01:00

9 lines
764 B
Plaintext

ggml_cuda_init: found 1 ROCm devices (Total VRAM: 126976 MiB):
Device 0: AMD Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32, VRAM: 126976 MiB
| model | size | params | backend | ngl | n_ubatch | fa | mmap | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | ---: | --------------: | -------------------: |
| llama 7B Q4_0 | 3.56 GiB | 6.74 B | ROCm | 99 | 2048 | 1 | 0 | pp2048 @ d32768 | 159.17 ± 0.71 |
| llama 7B Q4_0 | 3.56 GiB | 6.74 B | ROCm | 99 | 2048 | 1 | 0 | tg32 @ d32768 | 5.64 ± 0.00 |
build: 8e1f9d083 (9112)