benchmark/results/Qwen3-Next-80B-A3B-Instruct-UD-Q4_K_XL__rocm-7-nightly-rocwmma__hblt0__fa1__dual.log

ggml_cuda_init: GGML_CUDA_FORCE_MMQ:    no
ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no
ggml_cuda_init: found 2 ROCm devices:
  Device 0: AMD Radeon AI PRO R9700, gfx1201 (0x1201), VMM: no, Wave Size: 32
  Device 1: AMD Radeon AI PRO R9700, gfx1201 (0x1201), VMM: no, Wave Size: 32
| model                          |       size |     params | backend    | ngl | fa |            test |                  t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | --------------: | -------------------: |
| qwen3next 80B.A3B Q4_K - Medium |  42.01 GiB |    79.67 B | ROCm       |  99 |  1 |           pp512 |        937.81 ± 2.28 |
| qwen3next 80B.A3B Q4_K - Medium |  42.01 GiB |    79.67 B | ROCm       |  99 |  1 |           tg128 |         33.42 ± 0.01 |

build: 2aa45ef9e (7423)
added initial benchmark results 2025-11-30 11:17:29 +00:00			`ggml_cuda_init: GGML_CUDA_FORCE_MMQ: no`
			`ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no`
			`ggml_cuda_init: found 2 ROCm devices:`
			`Device 0: AMD Radeon AI PRO R9700, gfx1201 (0x1201), VMM: no, Wave Size: 32`
			`Device 1: AMD Radeon AI PRO R9700, gfx1201 (0x1201), VMM: no, Wave Size: 32`
			`\| model \| size \| params \| backend \| ngl \| fa \| test \| t/s \|`
			`\| ------------------------------ \| ---------: \| ---------: \| ---------- \| --: \| -: \| --------------: \| -------------------: \|`
benchmakr updates 2025-12-17 08:01:12 +00:00			`\| qwen3next 80B.A3B Q4_K - Medium \| 42.01 GiB \| 79.67 B \| ROCm \| 99 \| 1 \| pp512 \| 937.81 ± 2.28 \|`
			`\| qwen3next 80B.A3B Q4_K - Medium \| 42.01 GiB \| 79.67 B \| ROCm \| 99 \| 1 \| tg128 \| 33.42 ± 0.01 \|`
added initial benchmark results 2025-11-30 11:17:29 +00:00
benchmakr updates 2025-12-17 08:01:12 +00:00			`build: 2aa45ef9e (7423)`