Skip to content
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.

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

Fix multi-image and 2x speed improvements (DS-VL2) #157

Open
wants to merge 2 commits into
base: main
Choose a base branch
from

Conversation

Blaizzy
Copy link
Owner

@Blaizzy Blaizzy commented Dec 23, 2024

Summary

This PR introduces critical improvements and optimizations for multi-image inference, resulting in significant performance gains.

Key Updates

  1. Fixes Multi-Image Inference
    Resolved issues to ensure correct and efficient handling of multi-image inputs.

  2. Achieves ~2x Speedup
    Improved processing efficiency for both prompt handling and token generation, nearly doubling performance.

Performance Metrics

Single-Image Inference

Performance remains consistent and stable, as shown below:
Single Image

Multi-Image Inference

Significant improvements are evident in multi-image inference, as illustrated below:
Multi Image

@Blaizzy Blaizzy changed the title Fix multi-image and 2x speed improvements Fix multi-image and 2x speed improvements (DS-VL2) Dec 23, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

1 participant