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

feat: add benchmark results for rocm-7_2_1-pr21344 and update results metadata

Browse Source
This commit is contained in:
Donato Capitella
2026-04-15 11:39:10 +01:00
parent c2754a810a
commit 9707a15df7
41 changed files with 2566 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files
benchmark/results/Devstral-2-123B-Instruct-2512-UD-Q4_K_XL-00001-of-00002__rocm-7_2_1-pr21344__fa1.log
+8
View File
@@ -0,0 +1,8 @@
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 |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
| llama ?B Q4_K - Medium | 70.31 GiB | 125.03 B | ROCm | 99 | 1 | 0 | pp512 | 70.42 ± 0.04 |
| llama ?B Q4_K - Medium | 70.31 GiB | 125.03 B | ROCm | 99 | 1 | 0 | tg128 | 2.96 ± 0.00 |
build: 7957de9dc (8645)
benchmark/results/Devstral-2-123B-Instruct-2512-UD-Q4_K_XL-00001-of-00002__rocm-7_2_1-pr21344__fa1__longctx32768.log
+8
View File
@@ -0,0 +1,8 @@
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 | n_ubatch | fa | mmap | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | ---: | --------------: | -------------------: |
| llama ?B Q4_K - Medium | 70.31 GiB | 125.03 B | ROCm | 99 | 2048 | 1 | 0 | pp2048 @ d32768 | 16.77 ± 0.09 |
| llama ?B Q4_K - Medium | 70.31 GiB | 125.03 B | ROCm | 99 | 2048 | 1 | 0 | tg32 @ d32768 | 2.06 ± 0.01 |
build: 7957de9dc (8645)
benchmark/results/GLM-4.7-Flash-BF16-00001-of-00002__rocm-7_2_1-pr21344__fa1.log
+8
View File
@@ -0,0 +1,8 @@
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 |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
| deepseek2 30B.A3B BF16 | 55.79 GiB | 29.94 B | ROCm | 99 | 1 | 0 | pp512 | 403.52 ± 2.05 |
| deepseek2 30B.A3B BF16 | 55.79 GiB | 29.94 B | ROCm | 99 | 1 | 0 | tg128 | 21.37 ± 0.02 |
build: 7957de9dc (8645)
benchmark/results/GLM-4.7-Flash-BF16-00001-of-00002__rocm-7_2_1-pr21344__fa1__longctx32768.log
+8
View File
@@ -0,0 +1,8 @@
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 | n_ubatch | fa | mmap | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | ---: | --------------: | -------------------: |
| deepseek2 30B.A3B BF16 | 55.79 GiB | 29.94 B | ROCm | 99 | 2048 | 1 | 0 | pp2048 @ d32768 | 87.98 ± 0.28 |
| deepseek2 30B.A3B BF16 | 55.79 GiB | 29.94 B | ROCm | 99 | 2048 | 1 | 0 | tg32 @ d32768 | 15.34 ± 0.00 |
build: 7957de9dc (8645)
benchmark/results/GLM-4.7-Flash-UD-Q8_K_XL__rocm-7_2_1-pr21344__fa1.log
+8
View File
@@ -0,0 +1,8 @@
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 |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
| deepseek2 30B.A3B Q8_0 | 32.70 GiB | 29.94 B | ROCm | 99 | 1 | 0 | pp512 | 947.24 ± 201.85 |
| deepseek2 30B.A3B Q8_0 | 32.70 GiB | 29.94 B | ROCm | 99 | 1 | 0 | tg128 | 33.23 ± 0.00 |
build: 7957de9dc (8645)
benchmark/results/GLM-4.7-Flash-UD-Q8_K_XL__rocm-7_2_1-pr21344__fa1__longctx32768.log
+8
View File
@@ -0,0 +1,8 @@
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 | n_ubatch | fa | mmap | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | ---: | --------------: | -------------------: |
| deepseek2 30B.A3B Q8_0 | 32.70 GiB | 29.94 B | ROCm | 99 | 2048 | 1 | 0 | pp2048 @ d32768 | 91.69 ± 0.58 |
| deepseek2 30B.A3B Q8_0 | 32.70 GiB | 29.94 B | ROCm | 99 | 2048 | 1 | 0 | tg32 @ d32768 | 20.63 ± 0.00 |
build: 7957de9dc (8645)
benchmark/results/MiniMax-M2.7-UD-Q3_K_XL-00001-of-00004__rocm-7_2_1-pr21344__fa1.log
+8
View File
@@ -0,0 +1,8 @@
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 |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
| minimax-m2 230B.A10B Q3_K - Medium | 94.93 GiB | 228.69 B | ROCm | 99 | 1 | 0 | pp512 | 374.66 ± 1.20 |
| minimax-m2 230B.A10B Q3_K - Medium | 94.93 GiB | 228.69 B | ROCm | 99 | 1 | 0 | tg128 | 23.59 ± 0.01 |
build: 7957de9dc (8645)
benchmark/results/MiniMax-M2.7-UD-Q3_K_XL-00001-of-00004__rocm-7_2_1-pr21344__fa1__longctx32768.log
+8
View File
@@ -0,0 +1,8 @@
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 | n_ubatch | fa | mmap | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | ---: | --------------: | -------------------: |
| minimax-m2 230B.A10B Q3_K - Medium | 94.93 GiB | 228.69 B | ROCm | 99 | 2048 | 1 | 0 | pp2048 @ d32768 | 62.97 ± 0.05 |
| minimax-m2 230B.A10B Q3_K - Medium | 94.93 GiB | 228.69 B | ROCm | 99 | 2048 | 1 | 0 | tg32 @ d32768 | 6.36 ± 0.01 |
build: 7957de9dc (8645)
benchmark/results/Ministral-3-14B-Instruct-2512-BF16__rocm-7_2_1-pr21344__fa1.log
+8
View File
@@ -0,0 +1,8 @@
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 |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
| mistral3 14B BF16 | 25.16 GiB | 13.51 B | ROCm | 99 | 1 | 0 | pp512 | 731.49 ± 3.16 |
| mistral3 14B BF16 | 25.16 GiB | 13.51 B | ROCm | 99 | 1 | 0 | tg128 | 8.42 ± 0.00 |
build: 7957de9dc (8645)
benchmark/results/Ministral-3-14B-Instruct-2512-BF16__rocm-7_2_1-pr21344__fa1__longctx32768.log
+8
View File
@@ -0,0 +1,8 @@
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 | n_ubatch | fa | mmap | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | ---: | --------------: | -------------------: |
| mistral3 14B BF16 | 25.16 GiB | 13.51 B | ROCm | 99 | 2048 | 1 | 0 | pp2048 @ d32768 | 164.94 ± 0.77 |
| mistral3 14B BF16 | 25.16 GiB | 13.51 B | ROCm | 99 | 2048 | 1 | 0 | tg32 @ d32768 | 7.08 ± 0.00 |
build: 7957de9dc (8645)
benchmark/results/NVIDIA-Nemotron-3-Super-120B-A12B-UD-Q4_K_XL-00001-of-00003__rocm-7_2_1-pr21344__fa1.log
+8
View File
@@ -0,0 +1,8 @@
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 |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
| nemotron_h_moe 120B.A12B Q4_K - Medium | 78.02 GiB | 120.67 B | ROCm | 99 | 1 | 0 | pp512 | 334.32 ± 1.13 |
| nemotron_h_moe 120B.A12B Q4_K - Medium | 78.02 GiB | 120.67 B | ROCm | 99 | 1 | 0 | tg128 | 16.01 ± 0.03 |
build: 7957de9dc (8645)
benchmark/results/NVIDIA-Nemotron-3-Super-120B-A12B-UD-Q4_K_XL-00001-of-00003__rocm-7_2_1-pr21344__fa1__longctx32768.log
+8
View File
@@ -0,0 +1,8 @@
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 | n_ubatch | fa | mmap | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | ---: | --------------: | -------------------: |
| nemotron_h_moe 120B.A12B Q4_K - Medium | 78.02 GiB | 120.67 B | ROCm | 99 | 2048 | 1 | 0 | pp2048 @ d32768 | 308.76 ± 0.39 |
| nemotron_h_moe 120B.A12B Q4_K - Medium | 78.02 GiB | 120.67 B | ROCm | 99 | 2048 | 1 | 0 | tg32 @ d32768 | 15.60 ± 0.01 |
build: 7957de9dc (8645)
benchmark/results/Qwen3-Coder-30B-A3B-Instruct-Q4_K_M__rocm-7_2_1-pr21344__fa1.log
+8
View File
@@ -0,0 +1,8 @@
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 | 1504.86 ± 5.67 |
| qwen3moe 30B.A3B Q4_K - Medium | 17.35 GiB | 30.53 B | ROCm | 99 | 1 | 0 | tg128 | 67.86 ± 0.21 |
build: 7957de9dc (8645)
benchmark/results/Qwen3-Coder-30B-A3B-Instruct-Q4_K_M__rocm-7_2_1-pr21344__fa1__longctx32768.log
+8
View File
@@ -0,0 +1,8 @@
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 | n_ubatch | fa | mmap | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | ---: | --------------: | -------------------: |
| qwen3moe 30B.A3B Q4_K - Medium | 17.35 GiB | 30.53 B | ROCm | 99 | 2048 | 1 | 0 | pp2048 @ d32768 | 214.08 ± 0.06 |
| qwen3moe 30B.A3B Q4_K - Medium | 17.35 GiB | 30.53 B | ROCm | 99 | 2048 | 1 | 0 | tg32 @ d32768 | 33.54 ± 0.02 |
build: 7957de9dc (8645)
benchmark/results/Qwen3.5-122B-A10B-UD-Q5_K_XL-00001-of-00003__rocm-7_2_1-pr21344__fa1.log
+8
View File
@@ -0,0 +1,8 @@
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 |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
| qwen35moe 122B.A10B Q5_K - Medium | 85.60 GiB | 122.11 B | ROCm | 99 | 1 | 0 | pp512 | 404.64 ± 1.49 |
| qwen35moe 122B.A10B Q5_K - Medium | 85.60 GiB | 122.11 B | ROCm | 99 | 1 | 0 | tg128 | 19.61 ± 0.03 |
build: 7957de9dc (8645)
benchmark/results/Qwen3.5-122B-A10B-UD-Q5_K_XL-00001-of-00003__rocm-7_2_1-pr21344__fa1__longctx32768.log
+8
View File
@@ -0,0 +1,8 @@
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 | n_ubatch | fa | mmap | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | ---: | --------------: | -------------------: |
| qwen35moe 122B.A10B Q5_K - Medium | 85.60 GiB | 122.11 B | ROCm | 99 | 2048 | 1 | 0 | pp2048 @ d32768 | 251.55 ± 9.38 |
| qwen35moe 122B.A10B Q5_K - Medium | 85.60 GiB | 122.11 B | ROCm | 99 | 2048 | 1 | 0 | tg32 @ d32768 | 16.51 ± 0.01 |
build: 7957de9dc (8645)
benchmark/results/Qwen3.5-35B-A3B-BF16-00001-of-00002__rocm-7_2_1-pr21344__fa1.log
+8
View File
@@ -0,0 +1,8 @@
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 |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
| qwen35moe 35B.A3B BF16 | 64.60 GiB | 34.66 B | ROCm | 99 | 1 | 0 | pp512 | 511.62 ± 3.03 |
| qwen35moe 35B.A3B BF16 | 64.60 GiB | 34.66 B | ROCm | 99 | 1 | 0 | tg128 | 23.68 ± 0.00 |
build: 7957de9dc (8645)
benchmark/results/Qwen3.5-35B-A3B-BF16-00001-of-00002__rocm-7_2_1-pr21344__fa1__longctx32768.log
+8
View File
@@ -0,0 +1,8 @@
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 | n_ubatch | fa | mmap | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | ---: | --------------: | -------------------: |
| qwen35moe 35B.A3B BF16 | 64.60 GiB | 34.66 B | ROCm | 99 | 2048 | 1 | 0 | pp2048 @ d32768 | 381.40 ± 0.84 |
| qwen35moe 35B.A3B BF16 | 64.60 GiB | 34.66 B | ROCm | 99 | 2048 | 1 | 0 | tg32 @ d32768 | 19.86 ± 3.06 |
build: 7957de9dc (8645)
benchmark/results/Qwen3.5-35B-A3B-UD-Q4_K_XL__rocm-7_2_1-pr21344__fa1.log
+8
View File
@@ -0,0 +1,8 @@
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 |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
| qwen35moe 35B.A3B Q4_K - Medium | 20.70 GiB | 34.66 B | ROCm | 99 | 1 | 0 | pp512 | 1380.26 ± 14.08 |
| qwen35moe 35B.A3B Q4_K - Medium | 20.70 GiB | 34.66 B | ROCm | 99 | 1 | 0 | tg128 | 47.71 ± 0.01 |
build: 7957de9dc (8645)
benchmark/results/Qwen3.5-35B-A3B-UD-Q4_K_XL__rocm-7_2_1-pr21344__fa1__longctx32768.log
+8
View File
@@ -0,0 +1,8 @@
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 | n_ubatch | fa | mmap | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | ---: | --------------: | -------------------: |
| qwen35moe 35B.A3B Q4_K - Medium | 20.70 GiB | 34.66 B | ROCm | 99 | 2048 | 1 | 0 | pp2048 @ d32768 | 680.80 ± 1.05 |
| qwen35moe 35B.A3B Q4_K - Medium | 20.70 GiB | 34.66 B | ROCm | 99 | 2048 | 1 | 0 | tg32 @ d32768 | 40.13 ± 0.14 |
build: 7957de9dc (8645)
benchmark/results/Qwen3.5-35B-A3B-UD-Q8_K_XL__rocm-7_2_1-pr21344__fa1.log
+8
View File
@@ -0,0 +1,8 @@
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 |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
| qwen35moe 35B.A3B Q8_0 | 45.33 GiB | 34.66 B | ROCm | 99 | 1 | 0 | pp512 | 739.92 ± 29.08 |
| qwen35moe 35B.A3B Q8_0 | 45.33 GiB | 34.66 B | ROCm | 99 | 1 | 0 | tg128 | 29.61 ± 0.01 |
build: 7957de9dc (8645)
benchmark/results/Qwen3.5-35B-A3B-UD-Q8_K_XL__rocm-7_2_1-pr21344__fa1__longctx32768.log
+8
View File
@@ -0,0 +1,8 @@
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 | n_ubatch | fa | mmap | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | ---: | --------------: | -------------------: |
| qwen35moe 35B.A3B Q8_0 | 45.33 GiB | 34.66 B | ROCm | 99 | 2048 | 1 | 0 | pp2048 @ d32768 | 563.74 ± 0.42 |
| qwen35moe 35B.A3B Q8_0 | 45.33 GiB | 34.66 B | ROCm | 99 | 2048 | 1 | 0 | tg32 @ d32768 | 26.46 ± 0.01 |
build: 7957de9dc (8645)
benchmark/results/gemma-4-26B-A4B-it-BF16-00001-of-00002__rocm-7_2_1-pr21344__fa1.log
+8
View File
@@ -0,0 +1,8 @@
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 |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
| gemma4 ?B BF16 | 47.02 GiB | 25.23 B | ROCm | 99 | 1 | 0 | pp512 | 825.29 ± 28.88 |
| gemma4 ?B BF16 | 47.02 GiB | 25.23 B | ROCm | 99 | 1 | 0 | tg128 | 22.95 ± 0.00 |
build: 7957de9dc (8645)
benchmark/results/gemma-4-26B-A4B-it-BF16-00001-of-00002__rocm-7_2_1-pr21344__fa1__longctx32768.log
+8
View File
@@ -0,0 +1,8 @@
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 | n_ubatch | fa | mmap | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | ---: | --------------: | -------------------: |
| gemma4 ?B BF16 | 47.02 GiB | 25.23 B | ROCm | 99 | 2048 | 1 | 0 | pp2048 @ d32768 | 619.64 ± 9.10 |
| gemma4 ?B BF16 | 47.02 GiB | 25.23 B | ROCm | 99 | 2048 | 1 | 0 | tg32 @ d32768 | 20.37 ± 0.00 |
build: 7957de9dc (8645)
benchmark/results/gemma-4-26B-A4B-it-UD-Q4_K_XL__rocm-7_2_1-pr21344__fa1.log
+8
View File
@@ -0,0 +1,8 @@
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 |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
| gemma4 ?B Q4_K - Medium | 15.90 GiB | 25.23 B | ROCm | 99 | 1 | 0 | pp512 | 1615.54 ± 7.79 |
| gemma4 ?B Q4_K - Medium | 15.90 GiB | 25.23 B | ROCm | 99 | 1 | 0 | tg128 | 46.95 ± 0.08 |
build: 7957de9dc (8645)
benchmark/results/gemma-4-26B-A4B-it-UD-Q4_K_XL__rocm-7_2_1-pr21344__fa1__longctx32768.log
+8
View File
@@ -0,0 +1,8 @@
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 | n_ubatch | fa | mmap | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | ---: | --------------: | -------------------: |
| gemma4 ?B Q4_K - Medium | 15.90 GiB | 25.23 B | ROCm | 99 | 2048 | 1 | 0 | pp2048 @ d32768 | 692.91 ± 9.60 |
| gemma4 ?B Q4_K - Medium | 15.90 GiB | 25.23 B | ROCm | 99 | 2048 | 1 | 0 | tg32 @ d32768 | 37.13 ± 0.02 |
build: 7957de9dc (8645)
benchmark/results/gemma-4-26B-A4B-it-UD-Q8_K_XL__rocm-7_2_1-pr21344__fa1.log
+8
View File
@@ -0,0 +1,8 @@
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 |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
| gemma4 ?B Q8_0 | 25.94 GiB | 25.23 B | ROCm | 99 | 1 | 0 | pp512 | 1510.77 ± 39.47 |
| gemma4 ?B Q8_0 | 25.94 GiB | 25.23 B | ROCm | 99 | 1 | 0 | tg128 | 41.68 ± 0.02 |
build: 7957de9dc (8645)
benchmark/results/gemma-4-26B-A4B-it-UD-Q8_K_XL__rocm-7_2_1-pr21344__fa1__longctx32768.log
+8
View File
@@ -0,0 +1,8 @@
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 | n_ubatch | fa | mmap | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | ---: | --------------: | -------------------: |
| gemma4 ?B Q8_0 | 25.94 GiB | 25.23 B | ROCm | 99 | 2048 | 1 | 0 | pp2048 @ d32768 | 683.73 ± 6.54 |
| gemma4 ?B Q8_0 | 25.94 GiB | 25.23 B | ROCm | 99 | 2048 | 1 | 0 | tg32 @ d32768 | 33.71 ± 0.01 |
build: 7957de9dc (8645)
benchmark/results/gemma-4-31B-it-BF16-00001-of-00002__rocm-7_2_1-pr21344__fa1.log
+8
View File
@@ -0,0 +1,8 @@
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 |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
| gemma4 ?B BF16 | 57.18 GiB | 30.70 B | ROCm | 99 | 1 | 0 | pp512 | 374.46 ± 0.56 |
| gemma4 ?B BF16 | 57.18 GiB | 30.70 B | ROCm | 99 | 1 | 0 | tg128 | 3.40 ± 0.00 |
build: 7957de9dc (8645)
benchmark/results/gemma-4-31B-it-BF16-00001-of-00002__rocm-7_2_1-pr21344__fa1__longctx32768.log
+8
View File
@@ -0,0 +1,8 @@
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 | n_ubatch | fa | mmap | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | ---: | --------------: | -------------------: |
| gemma4 ?B BF16 | 57.18 GiB | 30.70 B | ROCm | 99 | 2048 | 1 | 0 | pp2048 @ d32768 | 153.71 ± 1.55 |
| gemma4 ?B BF16 | 57.18 GiB | 30.70 B | ROCm | 99 | 2048 | 1 | 0 | tg32 @ d32768 | 3.17 ± 0.00 |
build: 7957de9dc (8645)
benchmark/results/gemma-4-31B-it-UD-Q4_K_XL__rocm-7_2_1-pr21344__fa1.log
+8
View File
@@ -0,0 +1,8 @@
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 |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
| gemma4 ?B Q4_K - Medium | 17.46 GiB | 30.70 B | ROCm | 99 | 1 | 0 | pp512 | 293.56 ± 0.37 |
| gemma4 ?B Q4_K - Medium | 17.46 GiB | 30.70 B | ROCm | 99 | 1 | 0 | tg128 | 10.51 ± 0.00 |
build: 7957de9dc (8645)
benchmark/results/gemma-4-31B-it-UD-Q4_K_XL__rocm-7_2_1-pr21344__fa1__longctx32768.log
+8
View File
@@ -0,0 +1,8 @@
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 | n_ubatch | fa | mmap | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | ---: | --------------: | -------------------: |
| gemma4 ?B Q4_K - Medium | 17.46 GiB | 30.70 B | ROCm | 99 | 2048 | 1 | 0 | pp2048 @ d32768 | 136.26 ± 1.06 |
| gemma4 ?B Q4_K - Medium | 17.46 GiB | 30.70 B | ROCm | 99 | 2048 | 1 | 0 | tg32 @ d32768 | 8.11 ± 0.00 |
build: 7957de9dc (8645)
benchmark/results/gemma-4-31B-it-UD-Q8_K_XL__rocm-7_2_1-pr21344__fa1.log
+8
View File
@@ -0,0 +1,8 @@
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 |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
| gemma4 ?B Q8_0 | 32.60 GiB | 30.70 B | ROCm | 99 | 1 | 0 | pp512 | 282.75 ± 0.35 |
| gemma4 ?B Q8_0 | 32.60 GiB | 30.70 B | ROCm | 99 | 1 | 0 | tg128 | 6.16 ± 0.00 |
build: 7957de9dc (8645)
benchmark/results/gemma-4-31B-it-UD-Q8_K_XL__rocm-7_2_1-pr21344__fa1__longctx32768.log
+8
View File
@@ -0,0 +1,8 @@
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 | n_ubatch | fa | mmap | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | ---: | --------------: | -------------------: |
| gemma4 ?B Q8_0 | 32.60 GiB | 30.70 B | ROCm | 99 | 2048 | 1 | 0 | pp2048 @ d32768 | 132.30 ± 1.20 |
| gemma4 ?B Q8_0 | 32.60 GiB | 30.70 B | ROCm | 99 | 2048 | 1 | 0 | tg32 @ d32768 | 5.26 ± 0.00 |
build: 7957de9dc (8645)
benchmark/results/gpt-oss-120b-mxfp4-00001-of-00003__rocm-7_2_1-pr21344__fa1.log
+8
View File
@@ -0,0 +1,8 @@
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 |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
| gpt-oss 120B MXFP4 MoE | 59.02 GiB | 116.83 B | ROCm | 99 | 1 | 0 | pp512 | 954.68 ± 36.82 |
| gpt-oss 120B MXFP4 MoE | 59.02 GiB | 116.83 B | ROCm | 99 | 1 | 0 | tg128 | 50.97 ± 0.02 |
build: 7957de9dc (8645)
benchmark/results/gpt-oss-120b-mxfp4-00001-of-00003__rocm-7_2_1-pr21344__fa1__longctx32768.log
+8
View File
@@ -0,0 +1,8 @@
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 | n_ubatch | fa | mmap | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | ---: | --------------: | -------------------: |
| gpt-oss 120B MXFP4 MoE | 59.02 GiB | 116.83 B | ROCm | 99 | 2048 | 1 | 0 | pp2048 @ d32768 | 313.40 ± 10.22 |
| gpt-oss 120B MXFP4 MoE | 59.02 GiB | 116.83 B | ROCm | 99 | 2048 | 1 | 0 | tg32 @ d32768 | 35.98 ± 0.01 |
build: 7957de9dc (8645)
benchmark/results/gpt-oss-20b-mxfp4__rocm-7_2_1-pr21344__fa1.log
+8
View File
@@ -0,0 +1,8 @@
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 |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
| gpt-oss 20B MXFP4 MoE | 11.27 GiB | 20.91 B | ROCm | 99 | 1 | 0 | pp512 | 2028.68 ± 6.34 |
| gpt-oss 20B MXFP4 MoE | 11.27 GiB | 20.91 B | ROCm | 99 | 1 | 0 | tg128 | 72.58 ± 0.09 |
build: 7957de9dc (8645)
benchmark/results/gpt-oss-20b-mxfp4__rocm-7_2_1-pr21344__fa1__longctx32768.log
+8
View File
@@ -0,0 +1,8 @@
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 | n_ubatch | fa | mmap | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | ---: | --------------: | -------------------: |
| gpt-oss 20B MXFP4 MoE | 11.27 GiB | 20.91 B | ROCm | 99 | 2048 | 1 | 0 | pp2048 @ d32768 | 496.41 ± 2.25 |
| gpt-oss 20B MXFP4 MoE | 11.27 GiB | 20.91 B | ROCm | 99 | 2048 | 1 | 0 | tg32 @ d32768 | 51.87 ± 0.02 |
build: 7957de9dc (8645)
benchmark/results/llama-2-7b.Q4_0__rocm-7_2_1-pr21344__fa1.log
+8
View File
@@ -0,0 +1,8 @@
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 |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | ---: | --------------: | -------------------: |
| llama 7B Q4_0 | 3.56 GiB | 6.74 B | ROCm | 99 | 1 | 0 | pp512 | 1519.53 ± 5.33 |
| llama 7B Q4_0 | 3.56 GiB | 6.74 B | ROCm | 99 | 1 | 0 | tg128 | 50.57 ± 0.15 |
build: 7957de9dc (8645)
benchmark/results/llama-2-7b.Q4_0__rocm-7_2_1-pr21344__fa1__longctx32768.log
+8
View File
@@ -0,0 +1,8 @@
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 | n_ubatch | fa | mmap | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | ---: | --------------: | -------------------: |
| llama 7B Q4_0 | 3.56 GiB | 6.74 B | ROCm | 99 | 2048 | 1 | 0 | pp2048 @ d32768 | 164.68 ± 0.77 |
| llama 7B Q4_0 | 3.56 GiB | 6.74 B | ROCm | 99 | 2048 | 1 | 0 | tg32 @ d32768 | 5.64 ± 0.00 |
build: 7957de9dc (8645)