Commit 3791bdf
committed
feat: add standalone qwen3vl vision-encoder benchmark tool
tools/mtmd/qwen3vl-encoder: a small executable that runs only the Qwen3-VL
vision encoder (ViT + patch-merge projection) through ggml, with no libllama
dependency (it compiles the clip sources directly). It loads an mmproj gguf,
preprocesses an image, and times clip_image_batch_encode across CPU / Vulkan /
OpenCL, with a cosine-similarity check (--dump / --ref) to keep changes
lossless. This gives a fast edit -> build -> run loop for optimizing the shared
Vulkan/OpenCL/CPU kernels of the vision encoder without the heavy llama build.
See tools/mtmd/qwen3vl-encoder/standalone_loop.md for the local + Android +
Firebase workflow.1 parent dceb04c commit 3791bdf
3 files changed
Lines changed: 791 additions & 0 deletions
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
127 | 127 | | |
128 | 128 | | |
129 | 129 | | |
| 130 | + | |
| 131 | + | |
| 132 | + | |
| 133 | + | |
| 134 | + | |
| 135 | + | |
| 136 | + | |
| 137 | + | |
| 138 | + | |
| 139 | + | |
| 140 | + | |
| 141 | + | |
| 142 | + | |
| 143 | + | |
| 144 | + | |
| 145 | + | |
| 146 | + | |
| 147 | + | |
| 148 | + | |
| 149 | + | |
| 150 | + | |
| 151 | + | |
| 152 | + | |
| 153 | + | |
| 154 | + | |
| 155 | + | |
| 156 | + | |
| 157 | + | |
| 158 | + | |
| 159 | + | |
| 160 | + | |
| 161 | + | |
0 commit comments