Issues

• NaN miscompile: async-collective-fusion + overlap-compute-collective corrupt Wan2.2-I2V ulysses+splash attention on v6e (blocks ~5% DiT speedup)

  AI-Hypercomputer/maxdiffusion#429 · Filter by author ThomasNing opened on Jun 25, 2026
  8

• LTX-2.3 pipeline: native SIGSEGVs on tpu7x

  AI-Hypercomputer/maxdiffusion#423 · Filter by author tsvikas opened on Jun 21, 2026
  1

  

• Run wan2.2 model on GPU A100

  AI-Hypercomputer/maxdiffusion#412 · Filter by author kliuchevsky opened on May 20, 2026

• [Bug] MacOs local dev install fails on nvidia dependency

  AI-Hypercomputer/maxdiffusion#393 · Filter by author samos123 opened on Apr 28, 2026

• [Feature Request] Full training and LoRA support for Wan 2.1 and Wan 2.2

  AI-Hypercomputer/maxdiffusion#362 · Filter by author linwk20 opened on Mar 23, 2026
  3

• [Bug]: CI Workflow fails due to pyink version mismatch between setup.sh and GitHub Actions

  AI-Hypercomputer/maxdiffusion#343 · Filter by author Perseus14 opened on Mar 3, 2026
  1

  

• Does Wan2.2 Inference support parrallelim like TP/CP or some cache feature ?

  AI-Hypercomputer/maxdiffusion#318 · Filter by author pengchengneo opened on Jan 23, 2026
  2

• Difficulty finding documentation on using Flux Dev Fill for Google Cloud TPU infrastructure

  AI-Hypercomputer/maxdiffusion#279 · Filter by author hasan-muzak opened on Nov 10, 2025
  3

• Can I use IP-adapter plus model for sdxl pipeline?

  AI-Hypercomputer/maxdiffusion#161 · Filter by author Suprhimp opened on Apr 1, 2025
  1

• 3.11.6 test failed tests.schedulers.test_scheduler_flax.FlaxDDPMSchedulerTest.test_full_loop_no_noise

  AI-Hypercomputer/maxdiffusion#138 · Filter by author SamuelMarks opened on Dec 19, 2024
  7

• Are there some GPU ready configurations and related documentation?

  AI-Hypercomputer/maxdiffusion#101 · Filter by author mhmaqbool opened on Sep 4, 2024
  2