added rocm-7.2 benchmarks

This commit is contained in:
Donato Capitella
2026-01-23 15:11:13 +00:00
parent 9511598be4
commit d674531182
69 changed files with 4359 additions and 1 deletions
@@ -0,0 +1,8 @@
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 | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | --------------: | -------------------: |
| llama ?B Q4_K - Medium | 70.31 GiB | 125.03 B | ROCm | 99 | 1 | pp512 | 66.52 ± 7.27 |
| llama ?B Q4_K - Medium | 70.31 GiB | 125.03 B | ROCm | 99 | 1 | tg128 | 2.76 ± 0.08 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
ggml_cuda_init: found 1 ROCm devices:
Device 0: Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
| model | size | params | backend | ngl | n_ubatch | fa | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | --------------: | -------------------: |
| llama ?B Q4_K - Medium | 70.31 GiB | 125.03 B | ROCm | 99 | 2048 | 1 | pp2048 @ d32768 | 16.57 ± 0.05 |
| llama ?B Q4_K - Medium | 70.31 GiB | 125.03 B | ROCm | 99 | 2048 | 1 | tg32 @ d32768 | 2.08 ± 0.00 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
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 | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | --------------: | -------------------: |
| llama ?B Q4_K - Medium | 70.31 GiB | 125.03 B | ROCm | 99 | 1 | pp512 | 71.47 ± 0.20 |
| llama ?B Q4_K - Medium | 70.31 GiB | 125.03 B | ROCm | 99 | 1 | tg128 | 2.82 ± 0.07 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
ggml_cuda_init: found 1 ROCm devices:
Device 0: Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
| model | size | params | backend | ngl | n_ubatch | fa | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | --------------: | -------------------: |
| llama ?B Q4_K - Medium | 70.31 GiB | 125.03 B | ROCm | 99 | 2048 | 1 | pp2048 @ d32768 | 16.47 ± 0.03 |
| llama ?B Q4_K - Medium | 70.31 GiB | 125.03 B | ROCm | 99 | 2048 | 1 | tg32 @ d32768 | 2.06 ± 0.02 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
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 | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | --------------: | -------------------: |
| glm4moe 106B.A12B Q4_K - Medium | 68.01 GiB | 110.47 B | ROCm | 99 | 1 | pp512 | 133.48 ± 0.41 |
| glm4moe 106B.A12B Q4_K - Medium | 68.01 GiB | 110.47 B | ROCm | 99 | 1 | tg128 | 22.52 ± 0.00 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
ggml_cuda_init: found 1 ROCm devices:
Device 0: Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
| model | size | params | backend | ngl | n_ubatch | fa | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | --------------: | -------------------: |
| glm4moe 106B.A12B Q4_K - Medium | 68.01 GiB | 110.47 B | ROCm | 99 | 2048 | 1 | pp2048 @ d32768 | 34.18 ± 0.22 |
| glm4moe 106B.A12B Q4_K - Medium | 68.01 GiB | 110.47 B | ROCm | 99 | 2048 | 1 | tg32 @ d32768 | 8.12 ± 0.14 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
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 | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | --------------: | -------------------: |
| glm4moe 106B.A12B Q4_K - Medium | 68.01 GiB | 110.47 B | ROCm | 99 | 1 | pp512 | 181.28 ± 1.15 |
| glm4moe 106B.A12B Q4_K - Medium | 68.01 GiB | 110.47 B | ROCm | 99 | 1 | tg128 | 22.65 ± 0.06 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
ggml_cuda_init: found 1 ROCm devices:
Device 0: Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
| model | size | params | backend | ngl | n_ubatch | fa | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | --------------: | -------------------: |
| glm4moe 106B.A12B Q4_K - Medium | 68.01 GiB | 110.47 B | ROCm | 99 | 2048 | 1 | pp2048 @ d32768 | 35.28 ± 0.24 |
| glm4moe 106B.A12B Q4_K - Medium | 68.01 GiB | 110.47 B | ROCm | 99 | 2048 | 1 | tg32 @ d32768 | 7.97 ± 0.43 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
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 | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | --------------: | -------------------: |
| glm4moe 106B.A12B Q6_K | 94.57 GiB | 110.47 B | ROCm | 99 | 1 | pp512 | 56.12 ± 0.13 |
| glm4moe 106B.A12B Q6_K | 94.57 GiB | 110.47 B | ROCm | 99 | 1 | tg128 | 16.60 ± 0.04 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
ggml_cuda_init: found 1 ROCm devices:
Device 0: Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
| model | size | params | backend | ngl | n_ubatch | fa | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | --------------: | -------------------: |
| glm4moe 106B.A12B Q6_K | 94.57 GiB | 110.47 B | ROCm | 99 | 2048 | 1 | pp2048 @ d32768 | 22.87 ± 0.07 |
| glm4moe 106B.A12B Q6_K | 94.57 GiB | 110.47 B | ROCm | 99 | 2048 | 1 | tg32 @ d32768 | 6.89 ± 0.67 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
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 | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | --------------: | -------------------: |
| glm4moe 106B.A12B Q6_K | 94.57 GiB | 110.47 B | ROCm | 99 | 1 | pp512 | 105.50 ± 0.54 |
| glm4moe 106B.A12B Q6_K | 94.57 GiB | 110.47 B | ROCm | 99 | 1 | tg128 | 16.65 ± 0.01 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
ggml_cuda_init: found 1 ROCm devices:
Device 0: Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
| model | size | params | backend | ngl | n_ubatch | fa | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | --------------: | -------------------: |
| glm4moe 106B.A12B Q6_K | 94.57 GiB | 110.47 B | ROCm | 99 | 2048 | 1 | pp2048 @ d32768 | 32.93 ± 0.13 |
| glm4moe 106B.A12B Q6_K | 94.57 GiB | 110.47 B | ROCm | 99 | 2048 | 1 | tg32 @ d32768 | 7.24 ± 0.04 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
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 | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | --------------: | -------------------: |
| llama 70B Q8_0 | 75.65 GiB | 70.55 B | ROCm | 99 | 1 | pp512 | 49.50 ± 0.01 |
| llama 70B Q8_0 | 75.65 GiB | 70.55 B | ROCm | 99 | 1 | tg128 | 2.78 ± 0.00 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
ggml_cuda_init: found 1 ROCm devices:
Device 0: Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
| model | size | params | backend | ngl | n_ubatch | fa | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | --------------: | -------------------: |
| llama 70B Q8_0 | 75.65 GiB | 70.55 B | ROCm | 99 | 2048 | 1 | pp2048 @ d32768 | 27.87 ± 0.75 |
| llama 70B Q8_0 | 75.65 GiB | 70.55 B | ROCm | 99 | 2048 | 1 | tg32 @ d32768 | 2.42 ± 0.06 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
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 | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | --------------: | -------------------: |
| llama 70B Q8_0 | 75.65 GiB | 70.55 B | ROCm | 99 | 1 | pp512 | 49.27 ± 0.02 |
| llama 70B Q8_0 | 75.65 GiB | 70.55 B | ROCm | 99 | 1 | tg128 | 2.79 ± 0.00 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
ggml_cuda_init: found 1 ROCm devices:
Device 0: Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
| model | size | params | backend | ngl | n_ubatch | fa | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | --------------: | -------------------: |
| llama 70B Q8_0 | 75.65 GiB | 70.55 B | ROCm | 99 | 2048 | 1 | pp2048 @ d32768 | 28.46 ± 0.36 |
| llama 70B Q8_0 | 75.65 GiB | 70.55 B | ROCm | 99 | 2048 | 1 | tg32 @ d32768 | 2.44 ± 0.04 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
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 | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | --------------: | -------------------: |
| mistral3 14B BF16 | 25.16 GiB | 13.51 B | ROCm | 99 | 1 | pp512 | 797.13 ± 2.39 |
| mistral3 14B BF16 | 25.16 GiB | 13.51 B | ROCm | 99 | 1 | tg128 | 8.49 ± 0.00 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
ggml_cuda_init: found 1 ROCm devices:
Device 0: Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
| model | size | params | backend | ngl | n_ubatch | fa | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | --------------: | -------------------: |
| mistral3 14B BF16 | 25.16 GiB | 13.51 B | ROCm | 99 | 2048 | 1 | pp2048 @ d32768 | 146.47 ± 5.52 |
| mistral3 14B BF16 | 25.16 GiB | 13.51 B | ROCm | 99 | 2048 | 1 | tg32 @ d32768 | 7.11 ± 0.00 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
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 | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | --------------: | -------------------: |
| mistral3 14B BF16 | 25.16 GiB | 13.51 B | ROCm | 99 | 1 | pp512 | 803.39 ± 2.22 |
| mistral3 14B BF16 | 25.16 GiB | 13.51 B | ROCm | 99 | 1 | tg128 | 8.49 ± 0.00 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
ggml_cuda_init: found 1 ROCm devices:
Device 0: Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
| model | size | params | backend | ngl | n_ubatch | fa | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | --------------: | -------------------: |
| mistral3 14B BF16 | 25.16 GiB | 13.51 B | ROCm | 99 | 2048 | 1 | pp2048 @ d32768 | 152.56 ± 6.51 |
| mistral3 14B BF16 | 25.16 GiB | 13.51 B | ROCm | 99 | 2048 | 1 | tg32 @ d32768 | 7.11 ± 0.00 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
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 | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | --------------: | -------------------: |
| nemotron_h_moe 31B.A3.5B Q8_0 | 37.66 GiB | 31.58 B | ROCm | 99 | 1 | pp512 | 393.57 ± 2.37 |
| nemotron_h_moe 31B.A3.5B Q8_0 | 37.66 GiB | 31.58 B | ROCm | 99 | 1 | tg128 | 41.69 ± 0.01 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
ggml_cuda_init: found 1 ROCm devices:
Device 0: Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
| model | size | params | backend | ngl | n_ubatch | fa | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | --------------: | -------------------: |
| nemotron_h_moe 31B.A3.5B Q8_0 | 37.66 GiB | 31.58 B | ROCm | 99 | 2048 | 1 | pp2048 @ d32768 | 559.16 ± 0.99 |
| nemotron_h_moe 31B.A3.5B Q8_0 | 37.66 GiB | 31.58 B | ROCm | 99 | 2048 | 1 | tg32 @ d32768 | 39.74 ± 0.02 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
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 | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | --------------: | -------------------: |
| nemotron_h_moe 31B.A3.5B Q8_0 | 37.66 GiB | 31.58 B | ROCm | 99 | 1 | pp512 | 389.10 ± 3.02 |
| nemotron_h_moe 31B.A3.5B Q8_0 | 37.66 GiB | 31.58 B | ROCm | 99 | 1 | tg128 | 41.68 ± 0.00 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
ggml_cuda_init: found 1 ROCm devices:
Device 0: Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
| model | size | params | backend | ngl | n_ubatch | fa | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | --------------: | -------------------: |
| nemotron_h_moe 31B.A3.5B Q8_0 | 37.66 GiB | 31.58 B | ROCm | 99 | 2048 | 1 | pp2048 @ d32768 | 558.52 ± 1.33 |
| nemotron_h_moe 31B.A3.5B Q8_0 | 37.66 GiB | 31.58 B | ROCm | 99 | 2048 | 1 | tg32 @ d32768 | 39.73 ± 0.02 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
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 | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | --------------: | -------------------: |
| qwen3moe 235B.A22B Q3_K - Medium | 96.99 GiB | 235.09 B | ROCm | 99 | 1 | pp512 | 178.32 ± 26.83 |
| qwen3moe 235B.A22B Q3_K - Medium | 96.99 GiB | 235.09 B | ROCm | 99 | 1 | tg128 | 12.97 ± 0.98 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
ggml_cuda_init: found 1 ROCm devices:
Device 0: Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
| model | size | params | backend | ngl | n_ubatch | fa | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | --------------: | -------------------: |
| qwen3moe 235B.A22B Q3_K - Medium | 96.99 GiB | 235.09 B | ROCm | 99 | 2048 | 1 | pp2048 @ d32768 | 38.11 ± 0.36 |
| qwen3moe 235B.A22B Q3_K - Medium | 96.99 GiB | 235.09 B | ROCm | 99 | 2048 | 1 | tg32 @ d32768 | 9.90 ± 0.00 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
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 | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | --------------: | -------------------: |
| qwen3moe 235B.A22B Q3_K - Medium | 96.99 GiB | 235.09 B | ROCm | 99 | 1 | pp512 | 167.63 ± 28.76 |
| qwen3moe 235B.A22B Q3_K - Medium | 96.99 GiB | 235.09 B | ROCm | 99 | 1 | tg128 | 12.73 ± 0.74 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
ggml_cuda_init: found 1 ROCm devices:
Device 0: Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
| model | size | params | backend | ngl | n_ubatch | fa | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | --------------: | -------------------: |
| qwen3moe 235B.A22B Q3_K - Medium | 96.99 GiB | 235.09 B | ROCm | 99 | 2048 | 1 | pp2048 @ d32768 | 37.92 ± 0.33 |
| qwen3moe 235B.A22B Q3_K - Medium | 96.99 GiB | 235.09 B | ROCm | 99 | 2048 | 1 | tg32 @ d32768 | 9.78 ± 0.20 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
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 | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | --------------: | -------------------: |
| qwen3moe 30B.A3B BF16 | 56.89 GiB | 30.53 B | ROCm | 99 | 1 | pp512 | 484.09 ± 10.61 |
| qwen3moe 30B.A3B BF16 | 56.89 GiB | 30.53 B | ROCm | 99 | 1 | tg128 | 27.12 ± 0.16 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
ggml_cuda_init: found 1 ROCm devices:
Device 0: Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
| model | size | params | backend | ngl | n_ubatch | fa | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | --------------: | -------------------: |
| qwen3moe 30B.A3B BF16 | 56.89 GiB | 30.53 B | ROCm | 99 | 2048 | 1 | pp2048 @ d32768 | 162.38 ± 4.20 |
| qwen3moe 30B.A3B BF16 | 56.89 GiB | 30.53 B | ROCm | 99 | 2048 | 1 | tg32 @ d32768 | 18.71 ± 1.16 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
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 | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | --------------: | -------------------: |
| qwen3moe 30B.A3B BF16 | 56.89 GiB | 30.53 B | ROCm | 99 | 1 | pp512 | 492.54 ± 2.48 |
| qwen3moe 30B.A3B BF16 | 56.89 GiB | 30.53 B | ROCm | 99 | 1 | tg128 | 27.09 ± 0.01 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
ggml_cuda_init: found 1 ROCm devices:
Device 0: Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
| model | size | params | backend | ngl | n_ubatch | fa | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | --------------: | -------------------: |
| qwen3moe 30B.A3B BF16 | 56.89 GiB | 30.53 B | ROCm | 99 | 2048 | 1 | pp2048 @ d32768 | 161.72 ± 4.90 |
| qwen3moe 30B.A3B BF16 | 56.89 GiB | 30.53 B | ROCm | 99 | 2048 | 1 | tg32 @ d32768 | 19.36 ± 0.00 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
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 | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | --------------: | -------------------: |
| qwen3moe 30B.A3B Q6_K | 24.53 GiB | 30.53 B | ROCm | 99 | 1 | pp512 | 813.78 ± 5.52 |
| qwen3moe 30B.A3B Q6_K | 24.53 GiB | 30.53 B | ROCm | 99 | 1 | tg128 | 58.57 ± 0.01 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
ggml_cuda_init: found 1 ROCm devices:
Device 0: Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
| model | size | params | backend | ngl | n_ubatch | fa | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | --------------: | -------------------: |
| qwen3moe 30B.A3B Q6_K | 24.53 GiB | 30.53 B | ROCm | 99 | 2048 | 1 | pp2048 @ d32768 | 154.84 ± 3.34 |
| qwen3moe 30B.A3B Q6_K | 24.53 GiB | 30.53 B | ROCm | 99 | 2048 | 1 | tg32 @ d32768 | 31.32 ± 0.02 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
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 | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | --------------: | -------------------: |
| qwen3moe 30B.A3B Q6_K | 24.53 GiB | 30.53 B | ROCm | 99 | 1 | pp512 | 789.10 ± 47.98 |
| qwen3moe 30B.A3B Q6_K | 24.53 GiB | 30.53 B | ROCm | 99 | 1 | tg128 | 58.51 ± 0.01 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
ggml_cuda_init: found 1 ROCm devices:
Device 0: Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
| model | size | params | backend | ngl | n_ubatch | fa | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | --------------: | -------------------: |
| qwen3moe 30B.A3B Q6_K | 24.53 GiB | 30.53 B | ROCm | 99 | 2048 | 1 | pp2048 @ d32768 | 155.23 ± 3.28 |
| qwen3moe 30B.A3B Q6_K | 24.53 GiB | 30.53 B | ROCm | 99 | 2048 | 1 | tg32 @ d32768 | 31.27 ± 0.01 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
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 | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | --------------: | -------------------: |
| qwen3moe 30B.A3B Q4_K - Medium | 17.35 GiB | 30.53 B | ROCm | 99 | 1 | pp512 | 1207.32 ± 7.42 |
| qwen3moe 30B.A3B Q4_K - Medium | 17.35 GiB | 30.53 B | ROCm | 99 | 1 | tg128 | 71.51 ± 0.01 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
ggml_cuda_init: found 1 ROCm devices:
Device 0: Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
| model | size | params | backend | ngl | n_ubatch | fa | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | --------------: | -------------------: |
| qwen3moe 30B.A3B Q4_K - Medium | 17.35 GiB | 30.53 B | ROCm | 99 | 2048 | 1 | pp2048 @ d32768 | 161.97 ± 3.98 |
| qwen3moe 30B.A3B Q4_K - Medium | 17.35 GiB | 30.53 B | ROCm | 99 | 2048 | 1 | tg32 @ d32768 | 34.41 ± 0.01 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
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 | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | --------------: | -------------------: |
| qwen3moe 30B.A3B Q4_K - Medium | 17.35 GiB | 30.53 B | ROCm | 99 | 1 | pp512 | 1096.05 ± 129.46 |
| qwen3moe 30B.A3B Q4_K - Medium | 17.35 GiB | 30.53 B | ROCm | 99 | 1 | tg128 | 71.57 ± 0.01 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
ggml_cuda_init: found 1 ROCm devices:
Device 0: Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
| model | size | params | backend | ngl | n_ubatch | fa | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | --------------: | -------------------: |
| qwen3moe 30B.A3B Q4_K - Medium | 17.35 GiB | 30.53 B | ROCm | 99 | 2048 | 1 | pp2048 @ d32768 | 160.95 ± 3.41 |
| qwen3moe 30B.A3B Q4_K - Medium | 17.35 GiB | 30.53 B | ROCm | 99 | 2048 | 1 | tg32 @ d32768 | 34.37 ± 0.01 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
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 | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | --------------: | -------------------: |
| qwen3next 80B.A3B Q8_0 | 86.68 GiB | 79.67 B | ROCm | 99 | 1 | pp512 | 211.96 ± 2.48 |
| qwen3next 80B.A3B Q8_0 | 86.68 GiB | 79.67 B | ROCm | 99 | 1 | tg128 | 28.74 ± 0.49 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
ggml_cuda_init: found 1 ROCm devices:
Device 0: Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
| model | size | params | backend | ngl | n_ubatch | fa | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | --------------: | -------------------: |
| qwen3next 80B.A3B Q8_0 | 86.68 GiB | 79.67 B | ROCm | 99 | 2048 | 1 | pp2048 @ d32768 | 239.82 ± 0.51 |
| qwen3next 80B.A3B Q8_0 | 86.68 GiB | 79.67 B | ROCm | 99 | 2048 | 1 | tg32 @ d32768 | 24.76 ± 1.87 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
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 | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | --------------: | -------------------: |
| qwen3next 80B.A3B Q8_0 | 86.68 GiB | 79.67 B | ROCm | 99 | 1 | pp512 | 262.57 ± 3.77 |
| qwen3next 80B.A3B Q8_0 | 86.68 GiB | 79.67 B | ROCm | 99 | 1 | tg128 | 28.94 ± 0.05 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
ggml_cuda_init: found 1 ROCm devices:
Device 0: Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
| model | size | params | backend | ngl | n_ubatch | fa | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | --------------: | -------------------: |
| qwen3next 80B.A3B Q8_0 | 86.68 GiB | 79.67 B | ROCm | 99 | 2048 | 1 | pp2048 @ d32768 | 295.41 ± 0.37 |
| qwen3next 80B.A3B Q8_0 | 86.68 GiB | 79.67 B | ROCm | 99 | 2048 | 1 | tg32 @ d32768 | 23.85 ± 3.58 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
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 | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | --------------: | -------------------: |
| gemma3 12B Q8_0 | 13.40 GiB | 11.77 B | ROCm | 99 | 1 | pp512 | 323.22 ± 0.21 |
| gemma3 12B Q8_0 | 13.40 GiB | 11.77 B | ROCm | 99 | 1 | tg128 | 14.24 ± 0.00 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
ggml_cuda_init: found 1 ROCm devices:
Device 0: Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
| model | size | params | backend | ngl | n_ubatch | fa | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | --------------: | -------------------: |
| gemma3 12B Q8_0 | 13.40 GiB | 11.77 B | ROCm | 99 | 2048 | 1 | pp2048 @ d32768 | 231.95 ± 3.74 |
| gemma3 12B Q8_0 | 13.40 GiB | 11.77 B | ROCm | 99 | 2048 | 1 | tg32 @ d32768 | 11.66 ± 0.00 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
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 | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | --------------: | -------------------: |
| gemma3 12B Q8_0 | 13.40 GiB | 11.77 B | ROCm | 99 | 1 | pp512 | 324.04 ± 0.14 |
| gemma3 12B Q8_0 | 13.40 GiB | 11.77 B | ROCm | 99 | 1 | tg128 | 14.25 ± 0.00 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
ggml_cuda_init: found 1 ROCm devices:
Device 0: Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
| model | size | params | backend | ngl | n_ubatch | fa | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | --------------: | -------------------: |
| gemma3 12B Q8_0 | 13.40 GiB | 11.77 B | ROCm | 99 | 2048 | 1 | pp2048 @ d32768 | 230.54 ± 3.36 |
| gemma3 12B Q8_0 | 13.40 GiB | 11.77 B | ROCm | 99 | 2048 | 1 | tg32 @ d32768 | 11.66 ± 0.00 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
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 | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | --------------: | -------------------: |
| gemma3 27B BF16 | 50.31 GiB | 27.01 B | ROCm | 99 | 1 | pp512 | 430.73 ± 1.02 |
| gemma3 27B BF16 | 50.31 GiB | 27.01 B | ROCm | 99 | 1 | tg128 | 3.86 ± 0.00 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
ggml_cuda_init: found 1 ROCm devices:
Device 0: Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
| model | size | params | backend | ngl | n_ubatch | fa | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | --------------: | -------------------: |
| gemma3 27B BF16 | 50.31 GiB | 27.01 B | ROCm | 99 | 2048 | 1 | pp2048 @ d32768 | 193.46 ± 0.50 |
| gemma3 27B BF16 | 50.31 GiB | 27.01 B | ROCm | 99 | 2048 | 1 | tg32 @ d32768 | 3.69 ± 0.01 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
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 | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | --------------: | -------------------: |
| gemma3 27B BF16 | 50.31 GiB | 27.01 B | ROCm | 99 | 1 | pp512 | 525.55 ± 1.67 |
| gemma3 27B BF16 | 50.31 GiB | 27.01 B | ROCm | 99 | 1 | tg128 | 4.00 ± 0.00 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
ggml_cuda_init: found 1 ROCm devices:
Device 0: Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
| model | size | params | backend | ngl | n_ubatch | fa | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | --------------: | -------------------: |
| gemma3 27B BF16 | 50.31 GiB | 27.01 B | ROCm | 99 | 2048 | 1 | pp2048 @ d32768 | 202.49 ± 3.52 |
| gemma3 27B BF16 | 50.31 GiB | 27.01 B | ROCm | 99 | 2048 | 1 | tg32 @ d32768 | 3.72 ± 0.02 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
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 | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | --------------: | -------------------: |
| gemma3 4B Q3_K - Small | 1.80 GiB | 3.88 B | ROCm | 99 | 1 | pp512 | 2858.32 ± 17.99 |
| gemma3 4B Q3_K - Small | 1.80 GiB | 3.88 B | ROCm | 99 | 1 | tg128 | 84.57 ± 0.01 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
ggml_cuda_init: found 1 ROCm devices:
Device 0: Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
| model | size | params | backend | ngl | n_ubatch | fa | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | --------------: | -------------------: |
| gemma3 4B Q3_K - Small | 1.80 GiB | 3.88 B | ROCm | 99 | 2048 | 1 | pp2048 @ d32768 | 1182.57 ± 31.53 |
| gemma3 4B Q3_K - Small | 1.80 GiB | 3.88 B | ROCm | 99 | 2048 | 1 | tg32 @ d32768 | 61.59 ± 0.02 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
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 | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | --------------: | -------------------: |
| gemma3 4B Q3_K - Small | 1.80 GiB | 3.88 B | ROCm | 99 | 1 | pp512 | 2679.11 ± 228.92 |
| gemma3 4B Q3_K - Small | 1.80 GiB | 3.88 B | ROCm | 99 | 1 | tg128 | 70.08 ± 4.20 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
ggml_cuda_init: found 1 ROCm devices:
Device 0: Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
| model | size | params | backend | ngl | n_ubatch | fa | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | --------------: | -------------------: |
| gemma3 4B Q3_K - Small | 1.80 GiB | 3.88 B | ROCm | 99 | 2048 | 1 | pp2048 @ d32768 | 1138.90 ± 19.11 |
| gemma3 4B Q3_K - Small | 1.80 GiB | 3.88 B | ROCm | 99 | 2048 | 1 | tg32 @ d32768 | 61.59 ± 0.04 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
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 | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | --------------: | -------------------: |
| gpt-oss 120B MXFP4 MoE | 59.02 GiB | 116.83 B | ROCm | 99 | 1 | pp512 | 181.34 ± 1.51 |
| gpt-oss 120B MXFP4 MoE | 59.02 GiB | 116.83 B | ROCm | 99 | 1 | tg128 | 52.25 ± 0.01 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
ggml_cuda_init: found 1 ROCm devices:
Device 0: Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
| model | size | params | backend | ngl | n_ubatch | fa | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | --------------: | -------------------: |
| gpt-oss 120B MXFP4 MoE | 59.02 GiB | 116.83 B | ROCm | 99 | 2048 | 1 | pp2048 @ d32768 | 217.76 ± 0.23 |
| gpt-oss 120B MXFP4 MoE | 59.02 GiB | 116.83 B | ROCm | 99 | 2048 | 1 | tg32 @ d32768 | 36.60 ± 0.02 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
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 | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | --------------: | -------------------: |
| gpt-oss 120B MXFP4 MoE | 59.02 GiB | 116.83 B | ROCm | 99 | 1 | pp512 | 181.33 ± 1.42 |
| gpt-oss 120B MXFP4 MoE | 59.02 GiB | 116.83 B | ROCm | 99 | 1 | tg128 | 52.27 ± 0.01 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
ggml_cuda_init: found 1 ROCm devices:
Device 0: Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
| model | size | params | backend | ngl | n_ubatch | fa | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | --------------: | -------------------: |
| gpt-oss 120B MXFP4 MoE | 59.02 GiB | 116.83 B | ROCm | 99 | 2048 | 1 | pp2048 @ d32768 | 215.89 ± 4.16 |
| gpt-oss 120B MXFP4 MoE | 59.02 GiB | 116.83 B | ROCm | 99 | 2048 | 1 | tg32 @ d32768 | 36.64 ± 0.02 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
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 | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | --------------: | -------------------: |
| gpt-oss 20B MXFP4 MoE | 11.27 GiB | 20.91 B | ROCm | 99 | 1 | pp512 | 548.07 ± 6.87 |
| gpt-oss 20B MXFP4 MoE | 11.27 GiB | 20.91 B | ROCm | 99 | 1 | tg128 | 73.66 ± 0.02 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
ggml_cuda_init: found 1 ROCm devices:
Device 0: Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
| model | size | params | backend | ngl | n_ubatch | fa | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | --------------: | -------------------: |
| gpt-oss 20B MXFP4 MoE | 11.27 GiB | 20.91 B | ROCm | 99 | 2048 | 1 | pp2048 @ d32768 | 337.62 ± 3.36 |
| gpt-oss 20B MXFP4 MoE | 11.27 GiB | 20.91 B | ROCm | 99 | 2048 | 1 | tg32 @ d32768 | 52.54 ± 0.01 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
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 | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | --------------: | -------------------: |
| gpt-oss 20B MXFP4 MoE | 11.27 GiB | 20.91 B | ROCm | 99 | 1 | pp512 | 546.89 ± 7.03 |
| gpt-oss 20B MXFP4 MoE | 11.27 GiB | 20.91 B | ROCm | 99 | 1 | tg128 | 73.57 ± 0.01 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
ggml_cuda_init: found 1 ROCm devices:
Device 0: Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
| model | size | params | backend | ngl | n_ubatch | fa | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | --------------: | -------------------: |
| gpt-oss 20B MXFP4 MoE | 11.27 GiB | 20.91 B | ROCm | 99 | 2048 | 1 | pp2048 @ d32768 | 337.07 ± 4.50 |
| gpt-oss 20B MXFP4 MoE | 11.27 GiB | 20.91 B | ROCm | 99 | 2048 | 1 | tg32 @ d32768 | 52.54 ± 0.01 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
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 | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | --------------: | -------------------: |
| llama 7B Q4_0 | 3.56 GiB | 6.74 B | ROCm | 99 | 1 | pp512 | 549.23 ± 0.33 |
| llama 7B Q4_0 | 3.56 GiB | 6.74 B | ROCm | 99 | 1 | tg128 | 50.62 ± 0.01 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
ggml_cuda_init: found 1 ROCm devices:
Device 0: Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
| model | size | params | backend | ngl | n_ubatch | fa | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | --------------: | -------------------: |
| llama 7B Q4_0 | 3.56 GiB | 6.74 B | ROCm | 99 | 2048 | 1 | pp2048 @ d32768 | 139.39 ± 0.75 |
| llama 7B Q4_0 | 3.56 GiB | 6.74 B | ROCm | 99 | 2048 | 1 | tg32 @ d32768 | 5.59 ± 0.00 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
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 | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | --------------: | -------------------: |
| llama 7B Q4_0 | 3.56 GiB | 6.74 B | ROCm | 99 | 1 | pp512 | 549.22 ± 0.08 |
| llama 7B Q4_0 | 3.56 GiB | 6.74 B | ROCm | 99 | 1 | tg128 | 50.67 ± 0.01 |
build: a14b960bc (7816)
@@ -0,0 +1,8 @@
ggml_cuda_init: found 1 ROCm devices:
Device 0: Radeon 8060S Graphics, gfx1151 (0x1151), VMM: no, Wave Size: 32
| model | size | params | backend | ngl | n_ubatch | fa | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -------: | -: | --------------: | -------------------: |
| llama 7B Q4_0 | 3.56 GiB | 6.74 B | ROCm | 99 | 2048 | 1 | pp2048 @ d32768 | 137.30 ± 1.39 |
| llama 7B Q4_0 | 3.56 GiB | 6.74 B | ROCm | 99 | 2048 | 1 | tg32 @ d32768 | 5.60 ± 0.00 |
build: a14b960bc (7816)
+3815 -1
View File
File diff suppressed because it is too large Load Diff