adding raw benchmark results
This commit is contained in:
@@ -4,7 +4,7 @@ ggml_cuda_init: found 1 ROCm devices:
|
||||
Device 0: AMD Radeon Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
|
||||
| model | size | params | backend | ngl | mmap | test | t/s |
|
||||
| ------------------------------ | ---------: | ---------: | ---------- | --: | ---: | --------------: | -------------------: |
|
||||
| qwen3moe 30B.A3B BF16 | 56.89 GiB | 30.53 B | ROCm | 99 | 0 | pp512 | 151.25 ± 3.33 |
|
||||
| qwen3moe 30B.A3B BF16 | 56.89 GiB | 30.53 B | ROCm | 99 | 0 | tg128 | 23.80 ± 0.09 |
|
||||
| qwen3moe 30B.A3B BF16 | 56.89 GiB | 30.53 B | ROCm | 999 | 0 | pp512 | 153.49 ± 1.19 |
|
||||
| qwen3moe 30B.A3B BF16 | 56.89 GiB | 30.53 B | ROCm | 999 | 0 | tg128 | 24.52 ± 0.00 |
|
||||
|
||||
build: 66625a59 (6040)
|
||||
build: cd6983d5 (6119)
|
||||
|
||||
Reference in New Issue
Block a user