Skip to content

Conversation

jeffbolznv
Copy link
Collaborator

Primarily helps bigger GPUs. I think this helps some non-MoE models as well. Probably only helps the coopmat2 path which converts Y to fp16.

5090 before

Z:\github\jeffbolznv\llama.cpp\build\bin\RelWithDebInfo>llama-bench.exe -fa 1 -n 0 -p 512 -r 100 --prio 1 -m c:\models\Qwen_Qwen3-30B-A3B-Q2_K.gguf -m c:\models\\deepseek-v2-lite-safetensors\deepseek-v2-lite-Q4_K_M.gguf -m c:\models\gpt-oss-20b-mxfp4.gguf
ggml_vulkan: Found 1 Vulkan devices:
ggml_vulkan: 0 = NVIDIA GeForce RTX 5090 (NVIDIA) | uma: 0 | fp16: 1 | bf16: 1 | warp size: 32 | shared memory: 49152 | int dot: 1 | matrix cores: NV_coopmat2
| model                          |       size |     params | backend    | ngl | fa |            test |                  t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | --------------: | -------------------: |
| qwen3moe 30B.A3B Q2_K - Medium |  10.15 GiB |    30.53 B | Vulkan     |  99 |  1 |           pp512 |      6293.79 ± 64.66 |
| deepseek2 16B Q4_K - Medium    |   9.65 GiB |    15.71 B | Vulkan     |  99 |  1 |           pp512 |      8554.05 ± 90.98 |
| gpt-oss 20B MXFP4 MoE          |  11.27 GiB |    20.91 B | Vulkan     |  99 |  1 |           pp512 |     7622.73 ± 169.20 |

5090 after

Z:\github\jeffbolznv\llama.cpp\build\bin\RelWithDebInfo>llama-bench.exe -fa 1 -n 0 -p 512 -r 100 --prio 1 -m c:\models\Qwen_Qwen3-30B-A3B-Q2_K.gguf -m c:\models\\deepseek-v2-lite-safetensors\deepseek-v2-lite-Q4_K_M.gguf -m c:\models\gpt-oss-20b-mxfp4.gguf
ggml_vulkan: Found 1 Vulkan devices:
ggml_vulkan: 0 = NVIDIA GeForce RTX 5090 (NVIDIA) | uma: 0 | fp16: 1 | bf16: 1 | warp size: 32 | shared memory: 49152 | int dot: 1 | matrix cores: NV_coopmat2
| model                          |       size |     params | backend    | ngl | fa |            test |                  t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | --------------: | -------------------: |
| qwen3moe 30B.A3B Q2_K - Medium |  10.15 GiB |    30.53 B | Vulkan     |  99 |  1 |           pp512 |      6642.64 ± 69.59 |
| deepseek2 16B Q4_K - Medium    |   9.65 GiB |    15.71 B | Vulkan     |  99 |  1 |           pp512 |     8963.25 ± 168.48 |
| gpt-oss 20B MXFP4 MoE          |  11.27 GiB |    20.91 B | Vulkan     |  99 |  1 |           pp512 |     7637.93 ± 178.80 |

4070 before

Z:\github\jeffbolznv\llama.cpp\build\bin\RelWithDebInfo>llama-bench.exe -fa 1 -n 0 -p 512 -r 100 --prio 1 -m c:\models\Qwen_Qwen3-30B-A3B-Q2_K.gguf -m c:\models\\deepseek-v2-lite-safetensors\deepseek-v2-lite-Q4_K_M.gguf -m c:\models\gpt-oss-20b-mxfp4.gguf
ggml_vulkan: Found 1 Vulkan devices:
ggml_vulkan: 0 = NVIDIA GeForce RTX 4070 (NVIDIA) | uma: 0 | fp16: 1 | bf16: 1 | warp size: 32 | shared memory: 49152 | int dot: 1 | matrix cores: NV_coopmat2
| model                          |       size |     params | backend    | ngl | fa |            test |                  t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | --------------: | -------------------: |
| qwen3moe 30B.A3B Q2_K - Medium |  10.15 GiB |    30.53 B | Vulkan     |  99 |  1 |           pp512 |      2250.45 ± 21.68 |
| deepseek2 16B Q4_K - Medium    |   9.65 GiB |    15.71 B | Vulkan     |  99 |  1 |           pp512 |      2721.79 ± 18.53 |
| gpt-oss 20B MXFP4 MoE          |  11.27 GiB |    20.91 B | Vulkan     |  99 |  1 |           pp512 |      2991.03 ± 29.67 |

4070 after

Z:\github\jeffbolznv\llama.cpp\build\bin\RelWithDebInfo>llama-bench.exe -fa 1 -n 0 -p 512 -r 100 --prio 1 -m c:\models\Qwen_Qwen3-30B-A3B-Q2_K.gguf -m c:\models\\deepseek-v2-lite-safetensors\deepseek-v2-lite-Q4_K_M.gguf -m c:\models\gpt-oss-20b-mxfp4.gguf
ggml_vulkan: Found 1 Vulkan devices:
ggml_vulkan: 0 = NVIDIA GeForce RTX 4070 (NVIDIA) | uma: 0 | fp16: 1 | bf16: 1 | warp size: 32 | shared memory: 49152 | int dot: 1 | matrix cores: NV_coopmat2
| model                          |       size |     params | backend    | ngl | fa |            test |                  t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | --------------: | -------------------: |
| qwen3moe 30B.A3B Q2_K - Medium |  10.15 GiB |    30.53 B | Vulkan     |  99 |  1 |           pp512 |      2277.99 ± 21.86 |
| deepseek2 16B Q4_K - Medium    |   9.65 GiB |    15.71 B | Vulkan     |  99 |  1 |           pp512 |      2738.75 ± 14.72 |
| gpt-oss 20B MXFP4 MoE          |  11.27 GiB |    20.91 B | Vulkan     |  99 |  1 |           pp512 |      3001.78 ± 28.78 |

@jeffbolznv jeffbolznv requested a review from 0cc4m as a code owner August 24, 2025 17:43
@github-actions github-actions bot added Vulkan Issues specific to the Vulkan backend ggml changes relating to the ggml tensor library for machine learning labels Aug 24, 2025
@0cc4m
Copy link
Collaborator

0cc4m commented Aug 29, 2025

Sorry about the delay, I'm pretty busy again this week. I'll take care of the open PRs this weekend.

@0cc4m 0cc4m merged commit 696fccf into ggml-org:master Aug 30, 2025
48 checks passed
qnixsynapse pushed a commit to menloresearch/llama.cpp that referenced this pull request Aug 30, 2025
walidbr pushed a commit to walidbr/llama.cpp that referenced this pull request Sep 7, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

ggml changes relating to the ggml tensor library for machine learning Vulkan Issues specific to the Vulkan backend

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants