nikola/llama_cpp
1
0
Fork 0
Code Issues Pull Requests Actions 17 Packages Projects Releases Wiki Activity

mtmd: add MiMo v2.5 vision (#22883)

Browse Source 
* mimo-v2.5: vision support

* mimo-v2.5: use fused qkv for vision

* mimi-v2.5: fix f16 vision overflow

* mimo-v2.5: comment cleanups

* mimo-v2.5: Flash doesn't have mmproj
more cleanup
remember to use filter_tensors

* mimo-v2.5: fix trailing whitespace
This commit is contained in:
AesSedai
2026-05-12 11:11:14 +02:00
committed by GitHub
parent 78fbbc2c07
commit 4178259130
12 changed files with 460 additions and 23 deletions
Download Patch File Download Diff File Expand all files Collapse all files
tools/mtmd/models/models.h
+9
View File
@@ -33,6 +33,15 @@ struct clip_graph_qwen3vl : clip_graph {
ggml_cgraph * build() override;
};
struct clip_graph_mimovl : clip_graph {
clip_graph_mimovl(clip_ctx * ctx, const clip_image_f32 & img) : clip_graph(ctx, img) {}
ggml_cgraph * build() override;
// Force F32 mat-mul accumulation to avoid F16 overflow in the FFN down-proj
// when the mmproj is stored in F16 (the source weights are BF16; downcasting
// to F16 reduces dynamic range below the SwiGLU output magnitude on the last few layers).
ggml_tensor * build_mm(ggml_tensor * w, ggml_tensor * x) const override;
};
struct clip_graph_step3vl : clip_graph {
clip_graph_step3vl(clip_ctx * ctx, const clip_image_f32 & img) : clip_graph(ctx, img) {}
ggml_cgraph * build() override;