Commit Graph

Select branches

Hide Pull Requests

auto_convert

gemma_tenc

main

stable

Mono Color

• c09b4b799b Update pyproject.toml stable City 2026-01-12 14:02:13 +01:00
• 923d1d162d Merge branch 'main' into stable City 2026-01-12 13:58:56 +01:00
• 6ea2651e7d Only include metadata on new comfy versions main City 2026-01-12 13:55:28 +01:00
• 9ecc3c48e7 Add Gemma3 12B Support (#402) City 2026-01-12 13:38:42 +01:00
• 2e7f529b29 update loader.py jayn7 2026-01-12 07:09:08 +07:00
• 243a525df9 Update loader.py hpr 2026-01-12 02:40:44 +07:00
• 36b16deb09 Support tokenizer recreation from metadata jayn7 2026-01-12 02:37:57 +07:00
• 5a537ab64e Update loader.py hpr 2026-01-10 12:50:56 +07:00
• a4c94821cb Add Gemma3 12B Support jayn7 2026-01-10 03:20:31 +07:00
• c2e3b0adc3 Accept size-one ndarrays of any dimension from gguf reader (#394) Adam Niederer 2026-01-11 13:47:18 -05:00
• 58625e1cb6 Added support to load model config from Metadata. (#399) Vantage with AIandCity 2026-01-12 00:00:59 +05:30
• 795e45156e Use int64 for index in dequant City 2025-12-18 09:11:56 +01:00
• 8493db6481 Limit this fix to BF16 City 2025-12-18 09:03:26 +01:00
• 1b254b246a Merge branch 'main' into stable City 2025-12-17 16:01:44 +01:00
• a57094a3d3 Fix for scales being in BF16 City 2025-12-17 15:52:16 +01:00
• 01f8845bf3 Add VL version of Qwen3 4B as valid arch City 2025-12-01 01:32:43 +01:00
• c5c11dd511 Add VL version of Qwen3 4B as valid arch City 2025-12-01 01:32:43 +01:00
• 47a78ff7d8 Update pyproject.toml City 2025-11-29 22:40:55 +01:00
• f2b2066dec Merge branch 'main' into stable City 2025-11-29 22:35:07 +01:00
• 79379af338 Add tokenizer convert logic for Mistral City 2025-11-29 22:21:30 +01:00
• 0c9edef33b Improve invalid tenc file message City 2025-11-29 22:20:13 +01:00
• bff07287ab Update pyproject.toml City 2025-11-27 22:31:18 +01:00
• 3e66bcb77a Merge branch 'main' into stable City 2025-11-27 22:30:46 +01:00
• 4dfc025e4b Merge branch 'main' into auto_convert auto_convert City 2025-11-27 22:24:01 +01:00
• a0a0d2e120 Z-Image conversion support City 2025-11-27 22:19:02 +01:00
• ec7506d680 Flux2 specific filter City 2025-11-27 22:05:57 +01:00
• d3e14efda6 Hotfix for fallback City 2025-11-27 06:25:54 +01:00
• 6aa9866ef3 Add IQ4_NL and IQ4_XS support City 2025-11-27 05:17:04 +01:00
• ad82d2dc60 Update pyproject.toml City 2025-11-27 00:03:25 +01:00
• f57847df85 Merge branch 'main' into stable City 2025-11-27 00:03:09 +01:00
• fa48477de8 Qwen3-4B support for Z-Image City 2025-11-26 23:22:52 +01:00
• 5ade9afe98 Merge branch 'main' into auto_convert City 2025-11-25 23:46:01 +01:00
• bf8cb71be6 Initial attempt at gemma2 tokenizer gemma_tenc City 2025-11-08 19:34:56 +01:00
• d5ec55ee6f Update pyproject.toml City 2025-11-07 23:33:47 +01:00
• ee344c98fc Merge branch 'main' into stable City 2025-11-07 23:33:11 +01:00
• 02dac863ee add default named_modules_to_munmap variable (#357) rattus 2025-11-08 08:31:20 +10:00
• d4fbdb0139 Add type string for older mmproj files City 2025-11-04 07:04:11 +01:00
• 776e5b56e5 Update pyproject.toml City 2025-11-03 15:59:14 +01:00
• ca582bcb6d Merge branch 'main' into stable City 2025-11-03 15:58:26 +01:00
• 100c06cda8 nodes: mmap detach weights before they are pinned (#355) rattus 2025-11-04 00:57:29 +10:00
• be2a08330d Fix dequant_dtype handling (#334) blepping 2025-09-14 15:10:50 -06:00
• 6b778afdc9 Fix attempt to create a tensor with dtype "target" (#330) blepping 2025-09-08 23:07:45 -06:00
• e61c6980dc Shape fix City 2025-08-27 12:13:38 +02:00
• bfd9b39080 Blind fix for S2V City 2025-08-27 11:04:22 +02:00
• 10606d77d0 Update pyproject.toml City 2025-08-20 01:36:07 +02:00
• 5a8de41066 Merge branch 'main' into stable City 2025-08-20 01:35:39 +02:00
• d247022e3f Better mmproj detection logic City 2025-08-20 00:51:28 +02:00
• 0ca2aeb797 Qwen2-VL vision support City 2025-08-19 17:50:55 +02:00
• 7c2ceff6ed Keep first/last block in high precision for qwen image City 2025-08-05 16:17:51 +02:00
• 001ffcf0b5 Merge branch 'main' into auto_convert City 2025-08-05 15:46:09 +02:00
• df44c9d8dc Update pyproject.toml City 2025-08-05 15:36:53 +02:00
• bdb8f62fde Merge branch 'main' into stable City 2025-08-05 15:36:32 +02:00
• cf0573351a Qwen2-VL text model support City 2025-08-05 15:34:19 +02:00
• f4ab1b08fe Change ffn target City 2025-08-05 15:17:26 +02:00
• ad34e98355 Update pyproject.toml City 2025-08-05 14:31:01 +02:00
• bf846d54a8 Merge branch 'main' into stable City 2025-08-05 14:30:37 +02:00
• 16341ccfc0 Convert code for qwen image City 2025-08-05 14:27:52 +02:00
• 324c45ab0f Merge branch 'main' into auto_convert City 2025-08-05 12:55:10 +02:00
• 79ba1d4b17 Qwen image arch string City 2025-08-05 12:44:57 +02:00
• f1377afd9b Updated quantizing tool guide to compile with Visual Studio 2022 (#288) sfinktah 2025-08-05 20:44:09 +10:00
• b1350f6f5a Merge pull request #304 from spawner1145/main spawner 2025-08-05 18:42:40 +08:00
• c51a948233 Update pyproject.toml City 2025-06-15 01:02:28 +02:00
• 7be358d3b5 Merge branch 'main' into stable City 2025-06-15 01:01:52 +02:00
• b3ec875a68 Cosmos convert logic City 2025-06-15 00:09:07 +02:00
• 7b5f018d7e Cosmos convert logic City 2025-06-15 00:09:07 +02:00
• dba51d70f4 Merge branch 'main' into auto_convert City 2025-06-14 22:29:19 +02:00
• 4a6c6513bb Update pyproject.toml City 2025-05-09 05:37:16 +02:00
• 1126c12cb9 Update pyproject.toml City 2025-05-09 01:22:04 +02:00
• c6437f7211 ComfyUI Registry publish workflow City 2025-05-09 01:20:44 +02:00
• ee3de95597 Set initial version for registry City 2025-05-09 01:15:34 +02:00
• a2b75978fd Create pyproject.toml City 2025-05-09 01:08:35 +02:00
• 6570efec69 Better warning for incompatible models City 2025-05-07 05:13:08 +02:00
• 03842fab0a Remove old workaround City 2025-05-07 04:42:20 +02:00
• 3d673c5c09 Wan FLF2V convert logic City 2025-04-26 21:16:43 +02:00
• 54a4854e0c Allow embedding shape mismatch City 2025-04-22 15:50:05 +02:00
• 66c50234a4 Merge pull request #266 from city96/llama_oom_hotfix City 2025-04-21 23:30:02 +02:00
• 220f78ec70 Force recalculation of model size on clone City 2025-04-21 15:22:11 +02:00
• 9d282bcdfd LLaMA OOM hotfix City 2025-04-21 04:25:30 +02:00
• a355f8b5ea Skip unlink logic if load/offload dev match City 2025-04-18 23:53:12 +02:00
• 3564e95b26 Dynamic CLIP input types (#263) City 2025-04-18 23:48:53 +02:00
• ab19059d14 Small fix City 2025-04-18 00:10:49 +02:00
• 756b0b3b42 Merge branch 'main' into auto_convert City 2025-04-17 05:31:10 +02:00
• e29f526c68 Disable mixing scaled FP8 + GGUF City 2025-04-16 16:07:51 +02:00
• 47bec61475 HiDream text encoder support City 2025-04-16 06:10:28 +02:00
• 3605bfbead Keep ffn down for HiDream City 2025-04-16 02:22:31 +02:00
• 2ef5a66a9f HiDream support City 2025-04-16 01:43:15 +02:00
• 1c225ddb40 Initial auto convert script City 2025-04-15 01:33:10 +02:00
• 794fc28536 Clean up logging City 2025-04-13 23:01:15 +02:00
• cc478deb5a Refactor 5D tensor fix logic City 2025-04-12 05:41:26 +02:00
• 6de4bdba30 Merge pull request #216 from city96/convert_refactor_new City 2025-04-10 08:11:58 +02:00
• ca97e050b3 Update README.md City 2025-04-10 08:04:17 +02:00
• de33c984c1 Merge branch 'main' into convert_refactor_new City 2025-04-10 07:12:14 +02:00
• 624687939f Refactor 5D patch and save loop City 2025-04-10 07:12:10 +02:00
• 0c3bdf336e Enable more torch.compile optimization by reducing graph breaks (#243) Xiangxi Guo (Ryan) 2025-04-09 17:46:33 -07:00
• 862e4147c4 Make 5D tensor fix less fragile City 2025-04-03 00:38:46 +02:00
• c36b7236e6 Update README.md City 2025-03-30 18:46:09 +02:00
• f3b968e137 Merge branch 'main' into convert_refactor_new City 2025-03-30 18:43:44 +02:00
• bbd89ae547 Update .gitignore City 2025-03-30 18:42:17 +02:00
• 65a2bc20bb wan bf16 fix City 2025-03-28 04:21:26 +01:00
• 1c873edeb6 Expose strip prefix as function City 2025-03-22 20:40:24 +01:00