 Disty0
|
de5d4f0165
|
SDNQ fix sr not doing anything
|
2025-12-09 19:57:34 +03:00 |
|
|
Disty0
|
949ff04577
|
SDNQ fix fp16 mm with fp8 weights and improve stochastic rounding performance
|
2025-12-09 17:41:29 +03:00 |
|
 vladmandic
|
69f0d6bf5d
|
lint
Signed-off-by: vladmandic <mandic00@live.com>
|
2025-12-08 18:12:47 +01:00 |
|
|
Disty0
|
d4e2cbb826
|
SDNQ fix torch.compile always being active
|
2025-12-08 18:15:08 +03:00 |
|
|
Disty0
|
3ae7ecdbad
|
SDNQ fix quantization_device getting ignored on post load quant
|
2025-12-08 01:29:52 +03:00 |
|
|
Disty0
|
6e05a12a49
|
SDNQ post process pre-quants after load
|
2025-12-08 01:08:53 +03:00 |
|
|
Disty0
|
0835ca6f66
|
SDNQ add explicit model.quantization_method = QuantizationMethod.SDNQ
|
2025-12-08 00:46:40 +03:00 |
|
|
Disty0
|
7a6356f8eb
|
SDNQ fix transformers v5 and check for torch._dynamo.config.disable
|
2025-12-08 00:36:15 +03:00 |
|
|
Disty0
|
4f90054bf7
|
SDNQ transformers v5 support
|
2025-12-07 21:37:41 +03:00 |
|
|
Disty0
|
7aa1bfdc70
|
Add get_modules_to_not_convert from transformers v5
|
2025-12-02 01:01:51 +03:00 |
|
|
Disty0
|
d9bc31e7da
|
Cleanup
|
2025-11-29 01:46:04 +03:00 |
|
|
Disty0
|
01a0f6b356
|
Warn and disable quantized matmul if triton is not available
|
2025-11-29 01:34:54 +03:00 |
|
|
Disty0
|
3e52009a4f
|
SDNQ assert Triton for quantized matmul
|
2025-11-29 00:54:19 +03:00 |
|
|
Disty0
|
55cf627ac6
|
add version to sdnq
|
2025-11-28 00:45:24 +03:00 |
|
|
Disty0
|
368eb3103a
|
cleanup
|
2025-11-27 18:40:15 +03:00 |
|
|
Disty0
|
da3c439059
|
SDNQ fix _tied_weights_keys is dict case
|
2025-11-25 19:37:46 +03:00 |
|
|
Disty0
|
41ef28bb78
|
SDNQ don't divide group_size
|
2025-11-22 16:44:13 +03:00 |
|
|
Disty0
|
b6e9332cfe
|
SDNQ de-couple matmul dtype and add fp16 matmul
|
2025-11-22 02:16:20 +03:00 |
|
|
Disty0
|
5308630b3a
|
SDNQ use dequantize_fp32 with uint16 + torch_dtype = fp16
|
2025-11-18 23:53:27 +03:00 |
|
|
Disty0
|
49cd85d388
|
SDNQ add training related changes
|
2025-11-18 22:46:14 +03:00 |
|
|
Disty0
|
3fbfae5963
|
cleanup
|
2025-11-18 02:37:10 +03:00 |
|
|
Disty0
|
1745ed53f8
|
Refactor SDNQDequantizer
|
2025-11-18 01:42:58 +03:00 |
|
|
Disty0
|
3a4d7795d8
|
SDNQ fix weights_dtype getting overwritten on post load quant
|
2025-11-14 16:51:10 +03:00 |
|
|
Disty0
|
6f33ec3357
|
SDNQ use the model quant params instead of user settings on Lora
|
2025-11-10 00:12:38 +03:00 |
|
|
Disty0
|
93f28f07ac
|
Make SDNQ not depended on quantization_config.json and fix invalid quantization_config getting attached to the model on load
|
2025-11-07 18:11:21 +03:00 |
|
|
Disty0
|
76d699dc09
|
SDNQ add common keys
|
2025-10-31 00:21:54 +03:00 |
|
|
Vladimir Mandic
|
d43091f1fa
|
lint set minimum to py310 and update rules
Signed-off-by: Vladimir Mandic <mandic00@live.com>
|
2025-10-29 11:28:09 -04:00 |
|
|
Disty0
|
6c937c2747
|
Fix transformers using all the ram
|
2025-10-29 13:09:03 +03:00 |
|
|
Vladimir Mandic
|
bc775f0530
|
add wan asymmetric vae upscaler
Signed-off-by: Vladimir Mandic <mandic00@live.com>
|
2025-10-28 13:55:46 -04:00 |
|
|
Disty0
|
ef72edf18f
|
SDNQ improve svd and low bit matmul perf
|
2025-10-19 00:06:07 +03:00 |
|
|
Disty0
|
f12caf81f9
|
SDNQ skip bad layers on svd and fix svd with dequantize_fp32
|
2025-10-17 17:25:50 +03:00 |
|
|
Vladimir Mandic
|
4f336d3aab
|
linting
Signed-off-by: Vladimir Mandic <mandic00@live.com>
|
2025-10-16 19:39:05 -04:00 |
|
|
Disty0
|
2cf9938d97
|
SDNQ fix sdxl unet quant config not getting saved
|
2025-10-17 00:08:17 +03:00 |
|
|
Disty0
|
63aad89676
|
remove the unused state_dict arg
|
2025-10-16 16:29:23 +03:00 |
|
|
Vladimir Mandic
|
070edb20b0
|
update transformers and fix quant params
Signed-off-by: Vladimir Mandic <mandic00@live.com>
|
2025-10-16 09:21:20 -04:00 |
|
|
Disty0
|
b601f0d402
|
SDNQ expose svd_steps and update module skip keys
|
2025-10-14 00:15:09 +03:00 |
|
|
Disty0
|
d4d24214b3
|
SDNQ use a better way of loading pre quants and cleanup
|
2025-10-13 14:06:13 +03:00 |
|
|
Vladimir Mandic
|
2e4e741d47
|
seedvt2
Signed-off-by: Vladimir Mandic <mandic00@live.com>
|
2025-10-12 15:35:08 -04:00 |
|
|
Disty0
|
a376f89fd6
|
Add type checking to SDNQConfig
|
2025-10-12 01:02:47 +03:00 |
|
|
Disty0
|
9206d9443e
|
SDNQ add dequantize model
|
2025-10-12 00:00:53 +03:00 |
|
|
Disty0
|
f7286c90d5
|
SDNQ add native pre-quant loader support to from_pretrained
|
2025-10-11 16:19:11 +03:00 |
|
|
Disty0
|
6bc83bc296
|
Prevent accelerate from splitting Linear and Conv layers and causing device mismatch errors
|
2025-10-11 03:19:30 +03:00 |
|
|
Disty0
|
0f785880ee
|
SDNQ fix a singular bias not getting offloaded
|
2025-10-11 02:38:49 +03:00 |
|
|
Disty0
|
c7aba8589b
|
SDNQ fix Qwen loading
|
2025-10-11 00:05:09 +03:00 |
|
|
Disty0
|
35277a79d3
|
cleanup x3
|
2025-10-08 01:21:11 +03:00 |
|
|
Disty0
|
9c16e2234a
|
cleanup
|
2025-10-08 01:18:12 +03:00 |
|
|
Disty0
|
25303bb182
|
cleanup
|
2025-10-08 01:16:25 +03:00 |
|
|
Disty0
|
bdcd07f713
|
Add add_module_skip_keys to pre-load quant too
|
2025-10-08 01:11:40 +03:00 |
|
|
Disty0
|
7fdf400e8b
|
cleanup
|
2025-10-08 00:41:04 +03:00 |
|
|
Disty0
|
df03ea9ba8
|
SDNQ add sdnq_post_load_quant and update Qwen keys
|
2025-10-08 00:29:36 +03:00 |
|