Mislim, nije najgori #ai reyultat na mom računaru ali je daleko od dobrog

$ llama-bench -m qwen2.5-coder-3b-instruct-q6_k.gguf -t 3 --cpu-strict 1
ggml_vulkan: Found 1 Vulkan devices:
ggml_vulkan: 0 = Intel(R) HD Graphics 630 (KBL GT2) (Intel open-source Mesa driver) | uma: 1 | fp16: 1 | bf16: 0 | warp size: 32 | shared memory: 65536 | int dot: 0 | matrix cores: none
| model                          |       size |     params | backend    | ngl | threads | cpu_strict |            test |                  t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | ------: | ---------: | --------------: | -------------------: |
| qwen2 3B Q6_K                  |   2.60 GiB |     3.40 B | Vulkan     |  99 |       3 |          1 |           pp512 |         25.32 ± 0.01 |
| qwen2 3B Q6_K                  |   2.60 GiB |     3.40 B | Vulkan     |  99 |       3 |          1 |           tg128 |          4.51 ± 0.00 |

build: unknown (7709)