Papers/2608.28670
🧪 Test?View on arXiv

Memory-Efficient Training-Free Acceleration of Diffusion Transformers with BaryCache

Author1, Author2, Author3, Author4, Author5

image generationvideo generationaccelerationmemory efficiency
2608.28670
Builder Relevance
80%
1h ago

Abstract

This paper presents a method for accelerating Diffusion Transformers using a Barycentric Extrapolator to improve sampling speed while reducing memory usage.

Reality Card

Core Claim

The proposed method achieves up to 3.30x end-to-end sampling speedup compared to baseline DiT inference without increasing VRAM footprint.

Method / Result

3.30x end-to-end sampling speedup

Limitations

The method's performance may vary based on specific implementation details and hardware configurations.

Paper to code

Verified implementation resources so builders can test the paper’s claims instead of stopping at the abstract.

No verified implementation link has been attached yet. AIBuzzHub will keep this panel separate from unverified search results.
← Back to all papers