aboutsummaryrefslogtreecommitdiffstats
path: root/modules/sd_models.py
AgeCommit message (Collapse)AuthorLines
2024-01-06Execute model_loaded_callback after moving to target deviceNuullll-3/+3
2024-01-01Merge pull request #14145 from drhead/zero-terminal-snrAUTOMATIC1111-0/+6
Implement zero terminal SNR noise schedule option
2023-12-06Fix forced reloadKohaku-Blueleaf-1/+1
2023-12-02fix variabledrhead-1/+1
2023-12-02Create alphas_cumprod_original on full precision pathdrhead-0/+1
2023-12-02Ensure the cached weight will not be affectedKohaku-Blueleaf-2/+2
2023-12-02Merge branch 'dev' into test-fp8Kohaku-Blueleaf-4/+13
2023-12-01Add support for SD 2.1 Turbo, by converting the state dict from SGM to LDM ↵MrCheeze-4/+13
on load
2023-11-29Protect alphas_cumprod from downcastingdrhead-0/+5
2023-11-25Fix pre-fp8Kohaku-Blueleaf-1/+1
2023-11-21Option for using fp16 weight when apply loraKohaku-Blueleaf-3/+11
2023-11-19Use options instead of cmd_argsKohaku-Blueleaf-29/+32
2023-11-16Merge branch 'dev' into test-fp8Kohaku-Blueleaf-1/+4
2023-11-05more changes for #13865: fix formatting, rename the function, add comment ↵AUTOMATIC1111-1/+1
and add a readme entry
2023-11-05linterAUTOMATIC1111-2/+2
2023-11-05Merge branch 'dev' into masterAUTOMATIC1111-20/+30
2023-11-05Use devices.torch_gc() instead of empty_cache()Ritesh Gangnani-1/+0
2023-11-05Add SSD-1B as a supported modelRitesh Gangnani-2/+6
2023-10-28ManualCast for 10/16 series gpuKohaku-Blueleaf-9/+12
2023-10-25ignore mps for fp8Kohaku-Blueleaf-1/+3
2023-10-25Fix alphas cumprodKohaku-Blueleaf-1/+2
2023-10-25Fix alphas_cumprod dtypeKohaku-Blueleaf-0/+1
2023-10-25fp8 for TEKohaku-Blueleaf-0/+7
2023-10-24Fix lintKohaku-Blueleaf-1/+1
2023-10-24Add CPU fp8 supportKohaku-Blueleaf-4/+16
Since norm layer need fp32, I only convert the linear operation layer(conv2d/linear) And TE have some pytorch function not support bf16 amp in CPU. I add a condition to indicate if the autocast is for unet.
2023-10-19Add sdxl only argKohaku-Blueleaf-0/+3
2023-10-19Add fp8 for sd unetKohaku-Blueleaf-0/+3
2023-10-15repair unload sd checkpoint buttonAUTOMATIC1111-12/+1
2023-10-14use shallow copy for #13535AUTOMATIC1111-2/+1
2023-10-14Merge pull request #13535 from chu8129/devAUTOMATIC1111-4/+5
fix: checkpoints_loaded:{checkpoint:state_dict}, model.load_state_dict issue in dict value empty
2023-10-07reverstwangqiuwen-0/+2
2023-10-07upwangqiuwen-6/+5
2023-09-30Merge pull request #13139 from AUTOMATIC1111/ckpt-dir-path-separatorAUTOMATIC1111-2/+3
fix `--ckpt-dir` path separator and option use `short name` for checkpoint dropdown
2023-09-30add missing import, simplify code, use patches module for #13276AUTOMATIC1111-7/+12
2023-09-30Merge pull request #13276 from woweenie/patch-1AUTOMATIC1111-1/+14
patch DDPM.register_betas so that users can put given_betas in model yaml
2023-09-18fix王秋文/qwwang-0/+1
2023-09-15patch DDPM.register_betas so that users can put given_betas in model yamlwoweenie-1/+14
2023-09-15use dict[key]=model; did not update orderdict order, should use move to endqiuwen.wang-0/+1
2023-09-08parsing string to pathw-e-w-2/+3
2023-08-30keep order in list of checkpoints when loading model that doesn't have a ↵AUTOMATIC1111-1/+21
checksum
2023-08-30keep order in list of checkpoints when loading model that doesn't have a ↵AUTOMATIC1111-1/+21
checksum
2023-08-23set devices.dtype_unet correctlyAUTOMATIC1111-1/+2
2023-08-22add --medvram-sdxlAUTOMATIC1111-8/+8
2023-08-21Fix for consistency with shared.opts.sd_vae of UIUminosachi-0/+1
2023-08-20Change where VAE state are stored in modelUminosachi-2/+0
2023-08-20Change to access sd_model attribute with dotUminosachi-2/+2
2023-08-20Store base_vae and loaded_vae_file in sd_modelUminosachi-16/+8
2023-08-20Fix SD VAE switch error after model reuseUminosachi-2/+20
2023-08-17resolve the issue with loading fp16 checkpoints while using --no-halfAUTOMATIC1111-1/+4
2023-08-16send weights to target device instead of CPU memoryAUTOMATIC1111-1/+16
generated by cgit v1.2.3 (git 2.39.1) at 2025-10-07 22:26:42 +0000