monster/Qwen-Image-Layered-MRP-MLX
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Commit Graph

  • a255e4f0bb Add chunked saving for low memory, add pre-quantized models to README main pr302-head-a255e4f Zimeng Xiong 2025-12-22 08:46:28 -0800
  • 067ba25c7e feat: Introduce qwen-image-layered model saving Zimeng Xiong 2025-12-22 08:21:24 -0800
  • fe6936449d Add Qwen-Image-Layered support for image decomposition into RGBA layers Zimeng Xiong 2025-12-22 07:55:12 -0800
  • d70c8edb8c Add small README note on pre-quantized models on Huggingface filipstrand 2025-12-17 21:20:22 +0100
  • 0ade9a504c
         Fix/lora save bloat (#293) Filip Strand 2025-12-06 19:35:57 +0100
  • f45082f880
         Small LoRA improvements (#291) Filip Strand 2025-12-05 03:18:27 +0100
  • 5369d5adc6 Update VLM prompt to match filipstrand 2025-12-04 01:49:55 +0100
  • a404632d07
         Release 0.13.1 (#290) Filip Strand 2025-12-04 00:45:24 +0100
  • 88a541f329 Fix VLM error with transformers v.5 filipstrand 2025-12-03 23:30:16 +0100
  • ef5fce289a Update documentation temporarily for prerelease to enable v.5.0rc of transformers filipstrand 2025-12-03 23:01:21 +0100
  • c5f8b73d9d Fix import filipstrand 2025-12-03 22:50:02 +0100
  • 8e5609132c
         Release 0.13.0 (#289) Filip Strand 2025-12-03 22:40:44 +0100
  • 6551e9f6a1
         Update dependencies (#288) Filip Strand 2025-12-03 22:13:55 +0100
  • 37d202d0df
         Add support for Z-image Turbo & major improvements for weight loading and parameter resolution (#284) Filip Strand 2025-12-03 15:19:32 +0100
  • eb83175680
         Fix: Auto-download tokenizer files for FIBO-vlm when missing (#283) Filip Strand 2025-11-27 18:15:56 +0100
  • 38b1b538fe Fix trailing spaces after backslashes in shell examples filipstrand 2025-11-27 16:50:09 +0100
  • 2e28a987c7
         Release 0.12.0 (#282) Filip Strand 2025-11-27 16:30:37 +0100
  • c2c860f4de
         Add Bria FIBO support (#279) Filip Strand 2025-11-27 13:01:55 +0100
  • c3e72bd73c Update release date for v.0.11.1 filipstrand 2025-11-13 14:06:25 +0100
  • 2e92984125
         Release 0.11.1 (#278) Filip Strand 2025-11-13 01:05:29 +0100
  • d4005efac0
         Rename QwenImageEditPlus to QwenImageEdit (#277) Filip Strand 2025-11-11 17:07:36 +0100
  • bb5392ef2d
         Remove legacy QwenImageEdit model, keep only QwenImageEditPlus (#276) Filip Strand 2025-11-11 16:38:02 +0100
  • f9768b2e1c
         Qwen image edit (#274) Filip Strand 2025-11-11 16:25:58 +0100
  • ffce6ce149 Update release date filipstrand 2025-10-14 00:39:43 +0200
  • d314aa6889
         Release 0.11.0 (#271) Filip Strand 2025-10-14 00:34:27 +0200
  • 83354c3921
         Add XMP/IPTC metadata support (#267) Alessandro Rizzo 2025-10-12 14:45:28 +0200
  • 6947a23d6c
         Scheduler interface implementation. Implement pre-existing linear scheduler only. (#258) Anthony Wu 2025-10-12 03:20:40 -0700
  • 1278bafc53
         Support additional Qwen LoRA naming conventions (#270) Filip Strand 2025-10-11 16:00:25 +0200
  • 98e03a46f2
         Add Qwen Image support (#269) Filip Strand 2025-10-06 07:41:14 +0200
  • ef07b5186f
         Removing Claude CI agent, replacement coming soon (#265) Anthony Wu 2025-09-28 18:28:23 -0700
  • 8c2cefdd51
         Update tests/image_generation image comparison logic to allow similar images that are _close enough_ (#263) Anthony Wu 2025-09-08 07:49:54 -0700
  • 49af3504e6
         Fix: loading non-quantized saved models (#261) Anthony Wu 2025-09-08 05:30:33 -0700
  • 0d9af4f380
         Python 3.13 sentencepiece and torch unblocked (#259) Anthony Wu 2025-08-31 13:06:26 -0700
  • aca69703d3
         make mflux a namespace package (prepare for mflux.mcp) (#249) Anthony Wu 2025-08-23 06:21:46 -0700
  • e4315ae7f9 Allow "dev-krea" alias too filipstrand 2025-08-04 20:03:05 +0200
  • 127faf7f40 Update release date filipstrand 2025-08-04 09:01:58 +0200
  • 482a2d06c9
         Fix bug when loading in weights after a resumed training run (#245) Filip Strand 2025-08-03 12:46:31 +0200
  • 4f9af80956
         Small README update (#244) Filip Strand 2025-08-03 11:27:30 +0200
  • beee9b9509
         Add support for FLUX.1-Krea-dev model (#241) Filip Strand 2025-08-03 00:41:00 +0200
  • b8498c59db
         Link to 4-bit quantized Kontext model on HF (#240) Aarni Koskela 2025-07-30 18:25:57 +0300
  • df66510380 Sync lockfile filipstrand 2025-07-20 12:13:03 +0200
  • d4c6acc97a Set mflux version to 0.9.6 filipstrand 2025-07-20 12:04:02 +0200
  • d20f623581
         Proper: Revert back to mlx==0.26.1 (#239) Filip Strand 2025-07-20 12:01:01 +0200
  • eba6c64f60 Revert "Release 0.9.6 (#238)" filipstrand 2025-07-20 11:58:55 +0200
  • dc3a564b51
         Release 0.9.6 (#238) Filip Strand 2025-07-20 11:58:37 +0200
  • 3073b9feea Release 0.9.5 with import fixes filipstrand 2025-07-17 12:06:40 +0200
  • 7df35522f5 Fix imports after empty mflux init file filipstrand 2025-07-17 11:57:44 +0200
  • 7ab1e4b276 Fix PyPI upload and reorder release process filipstrand 2025-07-17 11:36:41 +0200
  • fd15a72cda Empty init file filipstrand 2025-07-17 11:29:44 +0200
  • 3d62d16fce Fix error filipstrand 2025-07-17 11:20:29 +0200
  • f8d6c52135 Fix module error filipstrand 2025-07-17 11:18:00 +0200
  • 3c5df450cc
         Release v.0.9.4 (#237) Filip Strand 2025-07-17 11:12:53 +0200
  • aceeafca8e
         Update CHANGELOG.md for 0.9.3 (#236) Anthony Wu 2025-07-08 20:25:52 -0700
  • 42bf9e3d4d
         Hot Fix: coldstart weight download (version 0.9.3) (#235) Anthony Wu 2025-07-08 20:19:34 -0700
  • 4fc63f535a
         Release 0.9.2 (#233) Filip Strand 2025-07-08 21:46:07 +0200
  • 842be6dde2
         🏗️ uv build backend update (#232) Anthony Wu 2025-07-06 23:22:49 -0700
  • a187bbbab0
         🗃️ Implement Local-first behavior to be resilient when HF/network is down (#226) Anthony Wu 2025-07-05 04:55:17 -0700
  • b86a5bae30
         Release 0.9.1 (#230) Filip Strand 2025-07-04 18:34:52 +0200
  • 676b30db14
         🎨 Inpaint tool improved - more shapes, eraser, undos (#225) Anthony Wu 2025-07-04 08:45:17 -0700
  • 11beed80df
         🧹 Cleanup/bump pre-commits; intro mypy config, quick type ignores/fixes (#221) Anthony Wu 2025-07-03 15:40:20 -0700
  • b250cdd922
         Release 0.9 (#222) Filip Strand 2025-06-28 00:03:01 +0200
  • 7388a856e5
         Add Flux1.Kontext[dev] support (#219) Filip Strand 2025-06-27 23:37:19 +0200
  • b26a1d66b3
         🖼️↗️ Add scale factor support for image upscaling dimensions (#215) Anthony Wu 2025-06-27 10:51:00 -0700
  • 0125103634
         ➡️➡️ Tab Tab for zsh: CLI completion generator (#208) Anthony Wu 2025-06-27 06:48:51 -0700
  • 6c3ba1e065
         🗂️ Cache Refactor - use platformdirs helper (#201) Anthony Wu 2025-06-27 04:59:43 -0700
  • 935d482ffb Update action permissions filipstrand 2025-06-14 11:29:53 +0200
  • 991fd18902 Update release action with Git configuration filipstrand 2025-06-14 11:27:17 +0200
  • ea4aca5850
         Update release script (#211) Filip Strand 2025-06-14 11:18:27 +0200
  • 9a20e83331
         Small additions before 0.8.0 (#209) Filip Strand 2025-06-14 10:02:04 +0200
  • dac08fdfdc Let stdin_prompt image gen tests be more lightweight filipstrand 2025-06-13 00:26:43 +0200
  • d2296ec197 Fix error in lora_layers causing lora training to fail filipstrand 2025-06-12 23:36:00 +0200
  • 4eb3f90fc9 Fix cli test filipstrand 2025-06-12 23:28:46 +0200
  • 847404748f
         Generalise in-context functionality (#203) Filip Strand 2025-06-12 09:01:07 +0200
  • 6a41329a45
         Delete ignore_pr.md included in #204 (#207) Anthony Wu 2025-06-11 00:04:49 -0700
  • c31326ff33
         Support prompt text via stdin. e.g. LLMs can generate prompts that directly feed into mflux. (#204) Anthony Wu 2025-06-10 22:31:46 -0700
  • 8c2ecd6364 Upload reference concept heatmaps for tests filipstrand 2025-06-04 22:35:05 +0200
  • 55dd6ee3e8 Reuse existing method in joint transformer filipstrand 2025-06-04 22:27:22 +0200
  • 674c44e30d Fix test for test_lora_args filipstrand 2025-06-04 21:47:26 +0200
  • 13bfb24964
         Concept Attention (#199) Filip Strand 2025-06-04 19:27:03 +0200
  • aaae64ad6e
         Bump MLX version (#200) Filip Strand 2025-06-04 16:35:40 +0200
  • d3e71328b2
         LORA_LIBRARY_PATH for Unix-style resource discovery (#198) Anthony Wu 2025-06-04 07:33:31 -0700
  • 0003bbe2cc
         Add Claude PR assistant GitHub Actions workflow (#197) Filip Strand 2025-05-24 13:50:44 +0200
  • fa56f33161
         Add upscaling support (#184) Filip Strand 2025-05-22 19:40:27 +0200
  • eff3812421
         Cleanup FluxFill mask args (#195) Filip Strand 2025-05-22 13:02:18 +0200
  • bc2d73d477
         DepthPro additional refactor (#193) Filip Strand 2025-05-20 20:20:59 +0200
  • 477af82248
         Remove PyTorch dependency for DepthPro (#192) Filip Strand 2025-05-20 19:32:22 +0200
  • d20f2833da
         🪫 Battery Saver - manage power over long batch operations (#189) Anthony Wu 2025-05-19 08:13:41 -0700
  • 4b56633e6c
         Add support for --prompt-file for dynamic prompt updates in large batches (#185) Anthony Wu 2025-05-19 08:10:52 -0700
  • 3b47c0b7e5
         Possibility to balance the Redux function (#190) Alessandro 2025-05-19 15:28:15 +0200
  • 8fb1eb4116
         Merge pull request #187 from filipstrand/depth-model-refactor Filip Strand 2025-05-10 13:45:27 +0200
  • 42b4fd6d90 Refactor DepthPro object filipstrand 2025-05-10 13:33:02 +0200
  • 4cb8a5b203 Better init pattern for depth model filipstrand 2025-05-10 13:20:45 +0200
  • 5f0f39f362 Update calls in MemorySaver filipstrand 2025-05-07 22:00:13 +0200
  • 393a60fb2b Small fix for LoRA scales filipstrand 2025-05-06 13:01:26 +0200
  • 63bf8710d1 v.0.7.1 filipstrand 2025-05-06 07:09:51 +0200
  • c820987efe
         Merge pull request #183 from anthonywu/in-context-allow-lora-composition Filip Strand 2025-05-05 10:50:15 +0200
  • 136b76de01 propose --save-full-image option Anthony Wu 2025-05-01 00:34:49 -0700
  • 219290541a nit: ignore uv sync's uv.lock Anthony Wu 2025-04-30 22:01:22 -0700
  • 8b828ced27 In-Context generation: allow lora composition Anthony Wu 2025-04-30 22:00:33 -0700
  • bf977e80a0
         Merge pull request #176 from filipstrand/remove-requirements.txt Filip Strand 2025-04-26 11:01:35 +0200