City
e61c6980dc
Shape fix
2025-08-27 12:13:38 +02:00
City
bfd9b39080
Blind fix for S2V
2025-08-27 11:04:22 +02:00
City
7c2ceff6ed
Keep first/last block in high precision for qwen image
2025-08-05 16:17:51 +02:00
City
001ffcf0b5
Merge branch 'main' into auto_convert
2025-08-05 15:46:09 +02:00
City
cf0573351a
Qwen2-VL text model support
2025-08-05 15:34:19 +02:00
City
f4ab1b08fe
Change ffn target
2025-08-05 15:17:26 +02:00
City
16341ccfc0
Convert code for qwen image
2025-08-05 14:27:52 +02:00
City
324c45ab0f
Merge branch 'main' into auto_convert
2025-08-05 12:55:10 +02:00
City
79ba1d4b17
Qwen image arch string
2025-08-05 12:44:57 +02:00
sfinktah
f1377afd9b
Updated quantizing tool guide to compile with Visual Studio 2022 ( #288 )
2025-08-05 12:44:09 +02:00
spawner
b1350f6f5a
Merge pull request #304 from spawner1145/main
...
support lumina-image-2.0
2025-08-05 12:42:40 +02:00
City
b3ec875a68
Cosmos convert logic
2025-06-15 00:09:45 +02:00
City
7b5f018d7e
Cosmos convert logic
2025-06-15 00:09:07 +02:00
City
dba51d70f4
Merge branch 'main' into auto_convert
2025-06-14 22:29:19 +02:00
City
a2b75978fd
Create pyproject.toml
2025-05-09 01:08:35 +02:00
City
6570efec69
Better warning for incompatible models
...
#256
2025-05-07 05:13:08 +02:00
City
03842fab0a
Remove old workaround
2025-05-07 04:42:20 +02:00
City
3d673c5c09
Wan FLF2V convert logic
...
Forgot to commit this, also fixes print -> logging
2025-04-26 21:16:43 +02:00
City
54a4854e0c
Allow embedding shape mismatch
...
#269
2025-04-22 15:50:05 +02:00
City
66c50234a4
Merge pull request #266 from city96/llama_oom_hotfix
...
- Apply umt5 hotfix of the token embed runtime cost being too large to llama as well
- Force model size to be re-calculated after the initial patcher is loaded to avoid issue with empty linear layers not being counted
2025-04-21 23:30:02 +02:00
City
220f78ec70
Force recalculation of model size on clone
2025-04-21 15:22:11 +02:00
City
9d282bcdfd
LLaMA OOM hotfix
2025-04-21 04:25:30 +02:00
City
a355f8b5ea
Skip unlink logic if load/offload dev match
...
i.e. when forcing it to run on CPU or highvram enabled
2025-04-18 23:53:12 +02:00
City
3564e95b26
Dynamic CLIP input types ( #263 )
2025-04-18 23:48:53 +02:00
City
ab19059d14
Small fix
2025-04-18 00:10:49 +02:00
City
756b0b3b42
Merge branch 'main' into auto_convert
2025-04-17 05:31:10 +02:00
City
e29f526c68
Disable mixing scaled FP8 + GGUF
2025-04-16 16:07:51 +02:00
City
47bec61475
HiDream text encoder support
...
LLaMA GGUF logic already works due to being used for HunYuan
2025-04-16 06:10:28 +02:00
City
3605bfbead
Keep ffn down for HiDream
2025-04-16 02:22:31 +02:00
City
2ef5a66a9f
HiDream support
...
#248
2025-04-16 01:43:15 +02:00
City
1c225ddb40
Initial auto convert script
2025-04-15 01:33:10 +02:00
City
794fc28536
Clean up logging
...
Also fixes numpy warning
2025-04-13 23:01:15 +02:00
City
cc478deb5a
Refactor 5D tensor fix logic
2025-04-12 05:41:26 +02:00
City
6de4bdba30
Merge pull request #216 from city96/convert_refactor_new
...
convert.py/patch refactor
2025-04-10 08:11:58 +02:00
City
ca97e050b3
Update README.md
2025-04-10 08:04:17 +02:00
City
de33c984c1
Merge branch 'main' into convert_refactor_new
2025-04-10 07:12:14 +02:00
City
624687939f
Refactor 5D patch and save loop
2025-04-10 07:12:10 +02:00
Xiangxi Guo (Ryan)
0c3bdf336e
Enable more torch.compile optimization by reducing graph breaks ( #243 )
...
As title, there were 2 fundamental sources of graph breaks:
1. using a pre-existing set in a `torch.compile` region, this is a
[known issue](https://github.com/pytorch/pytorch/issues/145761 ) in
PyTorch and the fix is WIP, but it's a lot easier to change the set
into a tuple here to fix the graph break.
2. using tensor subclass `GGMLTensor`. The full support of tensor
subclass has been gated for `torch.compile`, but with [a recent WIP
stack](https://github.com/pytorch/pytorch/pull/149792 ),
we'll be able to remove the gating and have `torch.compile` fully
support the `GGMLTensor` in this repo. The only required change here
is avoid changing the `__class__` attribute, and instead using
`torch.Tensor` to convert the subclass object back to Tensor.
2025-04-10 02:46:33 +02:00
City
862e4147c4
Make 5D tensor fix less fragile
2025-04-03 00:38:46 +02:00
City
c36b7236e6
Update README.md
2025-03-30 18:46:09 +02:00
City
f3b968e137
Merge branch 'main' into convert_refactor_new
2025-03-30 18:43:44 +02:00
City
bbd89ae547
Update .gitignore
2025-03-30 18:42:17 +02:00
City
65a2bc20bb
wan bf16 fix
2025-03-28 04:21:26 +01:00
City
1c873edeb6
Expose strip prefix as function
...
For external convert/detection logic
2025-03-22 20:40:24 +01:00
City
bc5223b0e3
Fix qtype logging on newer gguf-py
2025-03-16 20:14:33 +01:00
City
4aebf1bf2c
Add compat mode pig
...
closes #230
2025-03-16 19:38:40 +01:00
City
5aa6e98ad7
Merge pull request #224 from city96/tokenizer_convert
...
Sentencepiece tokenizer from gguf metadata for UMT5
2025-03-14 17:53:02 +01:00
City
ad69decc8d
Update requirements.txt
2025-03-14 17:43:05 +01:00
City
31dffa061b
Add error message for missing packages
2025-03-14 17:37:14 +01:00
City
296b0c1419
Clean up field logic
2025-03-14 17:11:47 +01:00
City