| model                          |       size |     params | backend    | ngl | type_k | type_v |  fa |            test |                  t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -----: | -----: | --: | --------------: | -------------------: |
| qwen35 27B Q8_0                |  26.62 GiB |    26.90 B | CUDA       | 999 |   q8_0 |   q8_0 |   1 |           pp512 |      3672.80 ± 16.36 |
| qwen35 27B Q8_0                |  26.62 GiB |    26.90 B | CUDA       | 999 |   q8_0 |   q8_0 |   1 |          pp4096 |       3590.66 ± 4.17 |
| qwen35 27B Q8_0                |  26.62 GiB |    26.90 B | CUDA       | 999 |   q8_0 |   q8_0 |   1 |           tg128 |         50.45 ± 0.04 |
| qwen35 27B Q8_0                |  26.62 GiB |    26.90 B | CUDA       | 999 |   q8_0 |   q8_0 |   1 |  pp512 @ d16384 |      3480.54 ± 53.77 |
| qwen35 27B Q8_0                |  26.62 GiB |    26.90 B | CUDA       | 999 |   q8_0 |   q8_0 |   1 | pp4096 @ d16384 |       3095.29 ± 6.23 |
| qwen35 27B Q8_0                |  26.62 GiB |    26.90 B | CUDA       | 999 |   q8_0 |   q8_0 |   1 |  tg128 @ d16384 |         48.76 ± 0.03 |

build: d2462f8f7 (9590)
