Skip to content

Actions: AI-Hypercomputer/maxdiffusion

Actions

Add Testgrid Link to PR

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
XLML.yml
225 workflow runs
225 workflow runs

Filter by Event

⌥ + click/return to exclude

Filter by Status

⌥ + click/return to exclude

Filter by Branch

⌥ + click/return to exclude

Filter by Actor

⌥ + click/return to exclude
Speed up WAN 2.2 checkpoint load and compile (no output change) Add Testgrid Link to PR #232: Pull request #438 opened by syhuang22
5s syhuang22:reduce-load-compile-clean
syhuang22:reduce-load-compile-clean
5s
Follow up fix custom flash block sizes fallback in Ulysses attention Add Testgrid Link to PR #231: Pull request #437 opened by ninatu
9s ninatu/fix_heads_per_tile2
ninatu/fix_heads_per_tile2
9s
Fixed-m softmax for pure-Ulysses WAN 2.2 attention (+ CFG batch-fold fix) Add Testgrid Link to PR #230: Pull request #436 opened by syhuang22
5s syhuang22:fixed-m-ulysses
syhuang22:fixed-m-ulysses
5s
(WIP) Cleaned Flux2 Klein Implementation, with benchmarking done on v6 TPU Add Testgrid Link to PR #229: Pull request #434 opened by amepas
7s flux2klein-onboarding
flux2klein-onboarding
7s
Add MagCache inference acceleration for Wan2.2 (T2V + I2V) Add Testgrid Link to PR #228: Pull request #433 opened by HadarIngonyama
8s HadarIngonyama:magcache_wan22_integration
HadarIngonyama:magcache_wan22_integration
8s
Optimize WAN pipelines: Reduce host overhead, hoist transposes, and fix Cache bugs Add Testgrid Link to PR #227: Pull request #432 opened by Perseus14
6s wan-cleanup
wan-cleanup
6s
Refactor custom_splash_attention for improved clarity Add Testgrid Link to PR #226: Pull request #431 opened by Perseus14
6s custom_attn_fix
custom_attn_fix
6s
[WAN 2.2] 2D Ring Attention with Custom Kernel Add Testgrid Link to PR #225: Pull request #430 opened by eltsai
8s 2d_ring_custom_kernel
2d_ring_custom_kernel
8s
feat: experimental two-phase (head-chunked) Ulysses all-to-all Add Testgrid Link to PR #224: Pull request #428 opened by csgoogle
12s sagarchapara/ulysses-two-phase
sagarchapara/ulysses-two-phase
12s
Fix get_dummy_wan_inputs for two-expert Wan2.2-I2V-A14B pipeline Add Testgrid Link to PR #223: Pull request #427 opened by ThomasNing
6s ThomasNing:fix/wan22-i2v-2.2-quant-dummy-inputs
ThomasNing:fix/wan22-i2v-2.2-quant-dummy-inputs
6s
fix(generate_wan): avoid process0-only gate for profiling run Add Testgrid Link to PR #222: Pull request #426 opened by edgexyz
11s edgexyz:main
edgexyz:main
11s
feat(ltx2): LTX-2.3 LoRA support (map gate_logits + prompt_adaln keys, fix example config to 22b) Add Testgrid Link to PR #221: Pull request #425 opened by tsvikas
9s tsvikas:ltx2-3-lora-support
tsvikas:ltx2-3-lora-support
9s
Fix LTX2 multihost JAX Array fetching error in LTX2 pipeline Add Testgrid Link to PR #220: Pull request #424 opened by prishajain1
6s ltx2_fix_multihost
ltx2_fix_multihost
6s
Perf(LTX2): Comprehensive XLA, Memory, and Transformer Code Quality Optimizations Add Testgrid Link to PR #219: Pull request #422 opened by Perseus14
7s ltx2-improvements
ltx2-improvements
7s
fix(ltx2): Fix VAE timing regression for large batch sizes Add Testgrid Link to PR #218: Pull request #421 opened by mbohlool
11s fix-vae-timing-regression
fix-vae-timing-regression
11s
Fix custom flash block sizes fallback in Ulysses attention Add Testgrid Link to PR #217: Pull request #420 opened by ninatu
14s ninatu/fix_heads_per_tile
ninatu/fix_heads_per_tile
14s
Refactor WAN loading logic: granular flags & boilerplate cleanup Add Testgrid Link to PR #216: Pull request #419 opened by ninatu
9s ninatu/wan_load_params
ninatu/wan_load_params
9s
Add Stable Diffusion 1.5 support with data-parallel inference Add Testgrid Link to PR #215: Pull request #418 opened by csgoogle
12s add-sd15-support
add-sd15-support
12s
refactor Flux transformer to use scanned blocks, dynamic checkpointing, and decoupled projections Add Testgrid Link to PR #214: Pull request #417 opened by prishajain1
8s prisha/flux_training
prisha/flux_training
8s
Fix: Plumb vmem limit and bkv_compute_in to the custom kernel Add Testgrid Link to PR #213: Pull request #416 opened by eltsai
8s plumb_vmem_limit
plumb_vmem_limit
8s
feat(ltx2): make run_text_encoder_on_tpu default False and dynamically load torchax Add Testgrid Link to PR #212: Pull request #415 opened by Perseus14
6s ltx2_tpu_encoder_disable
ltx2_tpu_encoder_disable
6s
feat(ltx2): implement centralized, configuration-driven logical sharding strategy for LTX-2 and LTX-2.3 Add Testgrid Link to PR #211: Pull request #414 opened by Perseus14
6s ltx2_sharding
ltx2_sharding
6s
add g++ compiler to base image and implement hardware-aware FeedForward sharding Add Testgrid Link to PR #210: Pull request #413 opened by prishajain1
10s prisha/img_fix
prisha/img_fix
10s
Optimize Wan VAE: Add Temporal Chunking, On-TPU Postprocessing, Static JIT Support, and Configurable Dtypes Add Testgrid Link to PR #209: Pull request #411 opened by Perseus14
9s wan_vae_opt
wan_vae_opt
9s
Fix NaN in LTX2 scheduler when dynamic shift saturates sigmas before shift_terminal Add Testgrid Link to PR #208: Pull request #410 opened by flyingblackshark
6s flyingblackshark:main
flyingblackshark:main
6s