Skip to content

v1.3.0: USP, 2D/3D Parallel, FP8 Blockwise, ...

Choose a tag to compare

View all tags
@DefTruth DefTruth released this 11 Mar 10:35
· 175 commits to main since this release
v1.3.0
1a9f717
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

v1.3.0 Major Release: USP, 2D/3D Parallel, FP8 Blockwise, ...

Cache-DiT v1.3.0 is a major release after v.1.2.0, the major changes incuding:

  • cache-dit-generate command line tool
  • Optimize VAE Parallel comm, use batched isend/irecv
  • 2D/3D Parallelism: Hybrid CP(USP) + TP, e.g, SP2 + TP2
  • Support USP (hybrid ulysses and ring attention)
  • New models support: GLM-Image, FLUX.2-Klein, Helios, FireRed-Image-Edit, and more.
  • Support pass a quantize_config to enable_cache API
  • Support load cache, parallelism and quantization config from yaml, docs
  • FP8 Blockwise dynamic quantization support
  • AMD GPUs support
  • ...

Full Changelog: v1.2.0...v1.3.0

Assets 2
Loading