Files
amd-strix-halo-toolboxes/benchmark/results/llama-2-7b.Q4_0__rocm7-nightlies-rocwmma__hblt0__fa1.log
T

9 lines
673 B
Plaintext

ggml_cuda_init: found 1 ROCm devices:
Device 0: Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
| model | size | params | backend | ngl | fa | mmap | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
| llama 7B Q4_0 | 3.56 GiB | 6.74 B | ROCm | 99 | 1 | 0 | pp512 | 670.47 ± 0.30 |
| llama 7B Q4_0 | 3.56 GiB | 6.74 B | ROCm | 99 | 1 | 0 | tg128 | 50.61 ± 0.01 |
build: 9c142e3a2 (7670)