139 Commits

Author SHA1 Message Date
Dity0 251945939b Add Lloyd-Max quantization with use_codebook 2026-08-10 22:51:32 +03:00
Dity0 712a13f1a0 Move SDNQ to upstream SDNQ repo 2026-08-10 12:13:10 +03:00
Vladimir Mandic a1df374773 add support for nunchaku-lite models and engine
Signed-off-by: Vladimir Mandic <mandic00@live.com>
2026-08-04 15:31:05 +02:00
Vladimir Mandic a83795273e prompt enhance and caption use llm-context
Signed-off-by: Vladimir Mandic <mandic00@live.com>
2026-08-04 14:21:53 +02:00
Disty0 75926e6fc4 Remove sdnq_use_quantized_matmul and use sdnq_quantize_matmul_mode instead 2026-07-15 18:45:30 +03:00
Vladimir Mandic 5072ea406f debounce width/height updates
Signed-off-by: Vladimir Mandic <mandic00@live.com>
2026-06-23 13:55:47 +02:00
Disty0 8214b044e3 SDNQ show a warining for no Triton instead and stop overriding matmul to false 2026-06-17 02:17:21 +03:00
QualiaRain 023c829179 fix model/vae: wrong quant_config var, getattr guards, remote-vae latent_copy shadowing + response handling
Co-Authored-By: Claude <noreply@anthropic.com>
2026-06-12 13:00:39 -04:00
Disty0 03fa90ff35 Add warning for and skip post-load quant of pre-quants 2026-06-07 18:02:28 +03:00
Vladimir Mandic 1f24513507 optimize sdnq quant-on-load and add platform stats
Signed-off-by: Vladimir Mandic <mandic00@live.com>
2026-06-07 11:43:46 +02:00
Vladimir Mandic ec2e37ee6e enhance automated testing
Signed-off-by: Vladimir Mandic <mandic00@live.com>
2026-06-03 09:27:34 +02:00
Vladimir Mandic b2223c7f8c nunchaku improvements
Signed-off-by: Vladimir Mandic <mandic00@live.com>
2026-05-18 19:33:16 +02:00
Disty0 c227d5fa10 SDNQ add Hadamard Rotation support 2026-05-18 17:06:11 +03:00
vladmandic 456d17fcd3 cleanup bnb references
Signed-off-by: vladmandic <mandic00@live.com>
2026-04-12 19:06:08 +02:00
Disty0 a16ff95f18 SDNQ add nn.Embedding quantization support and add Gemma4 keys 2026-04-10 18:46:58 +03:00
awsr ff247d8fd2 Merge branch 'dev' into RUF013 2026-03-24 07:12:51 -07:00
vladmandic 53839e464c remove legacy quant loaders
Signed-off-by: vladmandic <mandic00@live.com>
2026-03-24 15:01:07 +01:00
awsr 641321d7d2 RUF013 updates 2026-03-24 04:34:38 -07:00
Vladimir Mandic 494e4a7a7b fixes 2026-02-19 16:20:17 +01:00
Vladimir Mandic 4006b3e093 pylint 2026-02-19 11:36:20 +01:00
Vladimir Mandic d65a2d1ebc ruff lint 2026-02-19 11:13:44 +01:00
Vladimir Mandic e5c494f999 cleanup logger 2026-02-19 11:09:13 +01:00
Vladimir Mandic a3074baf8b unified logger 2026-02-19 09:46:42 +01:00
vladmandic 73b90c5228 redo all locales
Signed-off-by: vladmandic <mandic00@live.com>
2026-02-14 21:48:06 +01:00
CalamitousFelicitousness 935a4fcb03 feat: add Nunchaku group to reference
Replace manual Model/TE checkboxes in Quantization Settings with a
dedicated "Nunchaku" tab in the Extra Networks menu where users can
directly select nunchaku-quantized model variants. Detection is now
using a +nunchaku path marker for disambiguation.
2026-02-07 22:27:05 +00:00
vladmandic b3d65f4559 logging cleanup
Signed-off-by: vladmandic <mandic00@live.com>
2026-01-16 11:32:09 +01:00
Disty0 c2bc47e0c1 SDNQ expose Dyn quant on settings 2026-01-14 16:54:50 +03:00
Disty0 01a0f6b356 Warn and disable quantized matmul if triton is not available 2025-11-29 01:34:54 +03:00
Disty0 b6e9332cfe SDNQ de-couple matmul dtype and add fp16 matmul 2025-11-22 02:16:20 +03:00
Disty0 0b1e09091e cleanup 2025-11-19 02:38:21 +03:00
Vladimir Mandic f2835499b1 kanvas bindings
Signed-off-by: Vladimir Mandic <mandic00@live.com>
2025-11-07 12:21:48 -05:00
vladmandic 60ac82b191 add basic xpu gpu monitor
Signed-off-by: vladmandic <mandic00@live.com>
2025-10-26 18:55:54 -04:00
Disty0 b601f0d402 SDNQ expose svd_steps and update module skip keys 2025-10-14 00:15:09 +03:00
Vladimir Mandic 2e4e741d47 seedvt2
Signed-off-by: Vladimir Mandic <mandic00@live.com>
2025-10-12 15:35:08 -04:00
Disty0 df03ea9ba8 SDNQ add sdnq_post_load_quant and update Qwen keys 2025-10-08 00:29:36 +03:00
Vladimir Mandic 0092a8b86b add quantization_config for post-load
Signed-off-by: Vladimir Mandic <mandic00@live.com>
2025-10-07 10:14:31 -04:00
Disty0 9e52d0c1fb SDNQ add SVDQuant quantization method 2025-10-05 22:50:30 +03:00
Vladimir Mandic 70a2c209b1 cleanup
Signed-off-by: Vladimir Mandic <mandic00@live.com>
2025-09-28 20:22:25 -04:00
Disty0 a8b850adf4 move hf quantizer hijacks to sdnq 2025-09-12 20:54:44 +03:00
Vladimir Mandic 8ed04fb9a6 cleanup
Signed-off-by: Vladimir Mandic <mandic00@live.com>
2025-09-11 13:02:52 -04:00
Vladimir Mandic 7940217764 add models_not_to_quant option
Signed-off-by: Vladimir Mandic <mandic00@live.com>
2025-09-11 12:31:07 -04:00
Disty0 bc7c89c070 add typing to sdpa hijacks 2025-09-10 04:43:57 +03:00
Disty0 251a1ce3d9 fix modules_to_not_convert and modules_dtype_dict not resetting 2025-09-10 04:20:57 +03:00
Disty0 dcaeed360d SDNQ set return device to gpu with shuffle weights option 2025-09-08 18:56:29 +03:00
Vladimir Mandic 78c2a629b6 add experimental tensorrt quantization
Signed-off-by: Vladimir Mandic <mandic00@live.com>
2025-09-05 10:43:05 -04:00
Vladimir Mandic 2124ab6879 trt experiment
Signed-off-by: Vladimir Mandic <mandic00@live.com>
2025-09-04 10:53:32 -04:00
Disty0 e49814098e Add sdnq_modules_dtype_dict 2025-08-20 14:58:54 +03:00
Disty0 0946710662 Add sdnq_modules_to_not_convert to UI settings 2025-08-20 04:38:20 +03:00
Disty0 8ca74d0cd2 SDNQ rename unused param_name arg to op 2025-08-13 22:10:30 +03:00
Disty0 15cb8fe9f8 SDNQ add modules_dtype_dict and fix Qwen Image with quants less than 5 bits 2025-08-13 00:07:36 +03:00