Error: Model loading error: Model initialization failed: exception: access violation writing 0x0000000000000108

[mkat2]

clip_model_load: loaded meta data with 19 key-value pairs and 455 tensors from D:\comfyui.git\app\models\LLM\GGUF\MiniCPM-v-4\mmproj-model-f16.gguf
clip_model_load: Dumping metadata keys/values. Note: KV overrides do not apply in this output.
clip_model_load: - kv 0: general.architecture str = clip
clip_model_load: - kv 1: clip.has_text_encoder bool = false
clip_model_load: - kv 2: clip.has_vision_encoder bool = true
clip_model_load: - kv 3: clip.has_minicpmv_projector bool = true
clip_model_load: - kv 4: general.file_type u32 = 1
clip_model_load: - kv 5: general.description str = image encoder for MiniCPM-V
clip_model_load: - kv 6: clip.projector_type str = resampler
clip_model_load: - kv 7: clip.minicpmv_version i32 = 5
clip_model_load: - kv 8: clip.vision.image_size u32 = 448
clip_model_load: - kv 9: clip.vision.patch_size u32 = 14
clip_model_load: - kv 10: clip.vision.embedding_length u32 = 1152
clip_model_load: - kv 11: clip.vision.feed_forward_length u32 = 4304
clip_model_load: - kv 12: clip.vision.projection_dim u32 = 0
clip_model_load: - kv 13: clip.vision.attention.head_count u32 = 16
clip_model_load: - kv 14: clip.vision.attention.layer_norm_epsilon f32 = 0.000001
clip_model_load: - kv 15: clip.vision.block_count u32 = 27
clip_model_load: - kv 16: clip.vision.image_mean arr[f32,3] = [0.500000, 0.500000, 0.500000]
clip_model_load: - kv 17: clip.vision.image_std arr[f32,3] = [0.500000, 0.500000, 0.500000]
clip_model_load: - kv 18: clip.use_gelu bool = true
clip_model_load: - type f32: 285 tensors
clip_model_load: - type f16: 170 tensors
clip_model_load: CLIP using CPU backend
clip_model_load: params backend buffer size = 914.34 MB (455 tensors)
key clip.vision.image_grid_pinpoints not found in file
key clip.vision.mm_patch_merge_type not found in file
key clip.vision.image_crop_resolution not found in file
clip_image_build_graph: 448 448

any idea what is going on?

[1038lab]

It looks like the issue is caused by an invalid model parameter — specifically, projection_dim is set to 0, which is likely triggering a memory access error during model initialization.

• try Re-download the model file, as it may be corrupted or incompatible with the current version.
• If the issue persists, consider using a different version of the projector model that matches your MiniCPM-V version.

https://github.com/1038lab/ComfyUI-MiniCPM?tab=readme-ov-file#supported-models
Let us know if you need help finding a compatible model version.

[mkat2]

It looks like the issue is caused by an invalid model parameter — specifically, projection_dim is set to 0, which is likely triggering a memory access error during model initialization.

• try Re-download the model file, as it may be corrupted or incompatible with the current version.
• If the issue persists, consider using a different version of the projector model that matches your MiniCPM-V version.

https://github.com/1038lab/ComfyUI-MiniCPM?tab=readme-ov-file#supported-models Let us know if you need help finding a compatible model version.llama_cpp_python 0.2.89

I do got the MiniCPM-V-4-int4 transformer version working, but none of the GGUF work. I re-downloaded them a couple of times. What can be done to use the GGUF? All requirements match the versions or are higher... something with the llama_cpp_python ? or is there something wrong with the mmproj-model-f16.gguf ? (which downloads automatically) how can i get a different mmproj-model file to test?