mtmd: add min/max pixels gguf metadata (ggml-org#19273)

Author: ngxson

gguf-py/gguf/constants.py

@@ -284,6 +284,8 @@ class Clip:
     class ClipVision:
         PROJECTOR_TYPE      = "clip.vision.projector_type" # for mixed modality models
         IMAGE_SIZE          = "clip.vision.image_size"
+        IMAGE_MIN_PIXELS    = "clip.vision.image_min_pixels"
+        IMAGE_MAX_PIXELS    = "clip.vision.image_max_pixels"
         PREPROC_IMAGE_SIZE  = "clip.vision.preproc_image_size"
         PATCH_SIZE          = "clip.vision.patch_size"
         EMBEDDING_LENGTH    = "clip.vision.embedding_length"

gguf-py/gguf/gguf_writer.py

@@ -1113,6 +1113,12 @@ def add_vision_attention_layernorm_eps(self, value: float) -> None:
     def add_vision_image_size(self, value: int) -> None:
         self.add_uint32(Keys.ClipVision.IMAGE_SIZE, value)
 
+    def add_vision_max_pixels(self, value: int) -> None:
+        self.add_uint32(Keys.ClipVision.IMAGE_MAX_PIXELS, value)
+
+    def add_vision_min_pixels(self, value: int) -> None:
+        self.add_uint32(Keys.ClipVision.IMAGE_MIN_PIXELS, value)
+
     def add_vision_preproc_image_size(self, value: int) -> None:
         self.add_uint32(Keys.ClipVision.PREPROC_IMAGE_SIZE, value)
 

tools/mtmd/clip-impl.h

@@ -36,6 +36,8 @@
 // vision-specific
 #define KEY_VISION_PROJ_TYPE    "clip.vision.projector_type" // for models with mixed modalities
 #define KEY_IMAGE_SIZE          "clip.vision.image_size"
+#define KEY_IMAGE_MIN_PIXELS    "clip.vision.image_min_pixels"
+#define KEY_IMAGE_MAX_PIXELS    "clip.vision.image_max_pixels"
 #define KEY_PREPROC_IMAGE_SIZE  "clip.vision.preproc_image_size"
 #define KEY_PATCH_SIZE          "clip.vision.patch_size"
 #define KEY_IMAGE_MEAN          "clip.vision.image_mean"