sgl-project/sglang

[Feature] Implement Image Number Limitation for Image Processors

Open

#8,540 opened on Jul 29, 2025

View on GitHub
 (3 comments) (0 reactions) (0 assignees)Python (6,216 forks)auto 404
Multi-modalgood first issue

Repository metrics

Stars
 (28,442 stars)
PR merge metrics
 (Avg merge 2d 1h) (1,000 merged PRs in 30d)

Description

Checklist

Motivation

At present, there is no limitation on the number of images during the processing of mm data. It is necessary to impose such a limitation to prevent potential OOM and other problems.

(When using fast image processor, the GPU memory usage is roughly 1GiB for each 4K image)

Related resources

Implementation Steps:

  1. Define a constant, such as IMAGE_NUM_LIMITATION = 5, for each processor.
  2. Modify the baseprocessor to utilize this constant for request validation.
  3. Incorporate a server argument to enable modification of this numerical value.

In case you lack familiarity with the mm data processing pipeline, it is advisable to first refer to here

Contributor guide

Research direction
Study the provided multimodal request lifecycle tutorial to understand the existing image processing pipeline. Then implement the image number limitation constant and validation in the base processor.
Tech stack
python
Domain
backend
Issue type
Feature
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
3
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
Half day
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
Active
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
Clear
Prerequisites
GitPython
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
60