Fix MinicpmV model converter and clip to avoid using hardcode. #14750
+111
−56
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
ngxson
requested changes
Jul 24, 2025
sure. what branch will you suggest to use? shall I create a new one say 'minicpmv'? |
CISC
reviewed
Jul 28, 2025
| @@ -44,6 +44,8 @@ | |||
| #define KEY_WIN_ATTN_PATTERN "clip.vision.n_wa_pattern" | |||
| #define KEY_ATTN_WINDOW_SIZE "clip.vision.window_size" | |||
| #define KEY_MINICPMV_VERSION "clip.minicpmv_version" | |||
| #define KEY_MINICPMV_QUERY_NUM "clip.minicpmv_query_num" | |||
| #define KEY_MINICPMV_PROJ_DIM "clip.minicpmv_projection_dim" | |||
There was a problem hiding this comment.
Suggested change
| #define KEY_MINICPMV_PROJ_DIM "clip.minicpmv_projection_dim" |
| @@ -201,6 +201,8 @@ struct clip_hparams { | |||
| // legacy | |||
| bool has_llava_projector = false; | |||
| int minicpmv_version = 0; | |||
| int32_t minicpmv_query_num = 0; // MiniCPM-V query number | |||
| int32_t minicpmv_projection_dim = 0; // MiniCPM-V projection dimension | |||
There was a problem hiding this comment.
Suggested change
| int32_t minicpmv_projection_dim = 0; // MiniCPM-V projection dimension |
| @@ -2110,7 +2105,18 @@ struct clip_model_loader { | |||
| get_u32(KEY_PATCH_SIZE, hparams.patch_size); | |||
| get_u32(KEY_IMAGE_CROP_RESOLUTION, hparams.image_crop_resolution, false); | |||
| get_i32(KEY_MINICPMV_VERSION, hparams.minicpmv_version, false); // legacy | |||
|
|
|||
| get_u32(KEY_MINICPMV_QUERY_NUM, hparams.minicpmv_query_num, false); | |||
| get_u32(KEY_MINICPMV_PROJ_DIM, hparams.minicpmv_projection_dim, false); | |||
There was a problem hiding this comment.
Suggested change
| get_u32(KEY_MINICPMV_PROJ_DIM, hparams.minicpmv_projection_dim, false); |
| # Read config.json to get actual model configuration | ||
| config_path = os.path.join(dir_model, "config.json") | ||
| model_config = {} | ||
| if os.path.exists(config_path): |
There was a problem hiding this comment.
Suggested change
| if os.path.exists(config_path): | |
| if os.path.isfile(config_path): |
Comment on lines
+682
to
+684
| resampler_emb_dim = model_config.get("hidden_size", 0) if model_config else 0 | ||
| fout.add_uint32("clip.minicpmv_query_num", query_num) | ||
| fout.add_uint32("clip.minicpmv_projection_dim", resampler_emb_dim) |
There was a problem hiding this comment.
Suggested change
| resampler_emb_dim = model_config.get("hidden_size", 0) if model_config else 0 | |
| fout.add_uint32("clip.minicpmv_query_num", query_num) | |
| fout.add_uint32("clip.minicpmv_projection_dim", resampler_emb_dim) | |
| fout.add_uint32("clip.minicpmv_query_num", query_num) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Make sure to read the contributing guidelines before submitting a PR
Tested by llama-mtmd-cli with multiple minicpmv models.