julian/amd-strix-halo-toolboxes
1
0
Fork 0
Code Issues Pull Requests 2 Actions Packages Projects Releases Wiki Activity

Updated benchmarks

Browse Source
This commit is contained in:
Donato Capitella
2025-11-15 08:36:25 +00:00
parent 1d945f2c21
commit 67fb3a002b
664 changed files with 32551 additions and 15807 deletions
Download Patch File Download Diff File Expand all files Collapse all files
benchmark/results/gpt-oss-20b-F32__rocm7_rc-rocwmma__hblt0__fa1.log
+5 -5
View File
@@ -2,9 +2,9 @@ ggml_cuda_init: GGML_CUDA_FORCE_MMQ: no
ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no
ggml_cuda_init: found 1 ROCm devices:
Device 0: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
| model | size | params | backend | ngl | fa | mmap | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
| gpt-oss 20B BF16 | 38.97 GiB | 20.91 B | ROCm | 99 | 1 | 0 | pp512 | 1421.53 ± 7.06 |
| gpt-oss 20B BF16 | 38.97 GiB | 20.91 B | ROCm | 99 | 1 | 0 | tg128 | 27.32 ± 0.00 |
| model | size | params | backend | ngl | n_ubatch | fa | mmap | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | ---: | --------------: | -------------------: |
| gpt-oss 20B BF16 | 38.97 GiB | 20.91 B | ROCm | 99 | 2048 | 1 | 0 | pp512 | 1426.84 ± 4.41 |
| gpt-oss 20B BF16 | 38.97 GiB | 20.91 B | ROCm | 99 | 2048 | 1 | 0 | tg128 | 28.36 ± 0.00 |
build: a3cb0474 (6735)
build: bca95ca51 (7036)