| Name | Modified | Size | Downloads / Week |
|---|---|---|---|
| Parent folder | |||
| README.md | 2025-12-10 | 856 Bytes | |
| v2.5.19 source code.tar.gz | 2025-12-10 | 4.8 MB | |
| v2.5.19 source code.zip | 2025-12-10 | 4.8 MB | |
| Totals: 3 Items | 9.6 MB | 0 | |
- 🎨 New header logo design - Refreshed ASCII art banner (thanks @naxci1)
- 🧹 Remove dead flash attention wrapper - Removed legacy code from FP8CompatibleDiT; FlashAttentionVarlen already handles backend switching via its attention_mode attribute
- 🛡️ Fix graceful fallback from flash-attn - Add compatibility shims for corrupted flash_attn/xformers DLLs, preventing startup crashes when CUDA extensions are broken
- 📊 Improved VRAM tracking - Separate allocated vs reserved memory tracking, Windows-only overflow detection (WDDM paging behavior)
- ♻️ Centralize backend detection - Unified is_mps_available(), is_cuda_available(), get_gpu_backend() helpers across codebase
- 🔄 Revert 2.5.14 VRAM limit enforcement - Removed set_per_process_memory_fraction call; Overflow detection and warnings remain.