Skip to content

Update limit_mm_per_prompt format #2919

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
xyang16 merged 1 commit into from
Oct 23, 2025
Merged

Update limit_mm_per_prompt format #2919

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion engines/python/setup/djl_python/tests/test_properties_manager.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -618,7 +618,7 @@ def test_all_vllm_engine_args():
"tokenizer_pool_size": "12",
"tokenizer_pool_type": "mytype",
"tokenizer_pool_extra_config": '{"a": "b"}',
"limit_mm_per_prompt": "image=2",
"limit_mm_per_prompt": '{"image":2}',
"enable_lora": "true",
"max_loras": "5",
"max_lora_rank": "123",
Expand Down
4 changes: 2 additions & 2 deletions serving/docs/lmi/user_guides/vision_language_models.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -22,7 +22,7 @@ You can read more about the supported format in the [chat completions doc](chat_
Deploying Vision Language Models with LMI is very similar to deploying Text Generation Models.

There are some additional, optional configs that are exposed:
* `option.limit_mm_per_prompt`: For each multimodal plugin, limit how many input instances to allow for each prompt. Expects a comma-separated list of items, e.g.: `image=16,video=2` allows a maximum of 16 images and 2 videos per prompt. Defaults to 1 for each modality.
* `option.limit_mm_per_prompt`: For each multimodal plugin, limit how many input instances to allow for each prompt. Expects a comma-separated list of items, e.g.: `{"image": 16, "video": 2}` allows a maximum of 16 images and 2 videos per prompt. Defaults to 1 for each modality.

Example SageMaker deployment code:

Expand All @@ -32,7 +32,7 @@ from sagemaker.djl_inference import DJLModel
model = DJLModel(
model_id="llava-hf/llava-v1.6-mistral-7b-hf",
env={
"OPTION_LIMIT_MM_PER_PROMPT": "image=2",
"OPTION_LIMIT_MM_PER_PROMPT": '{"image":2}',
}
)

Expand Down
Loading