{"id":21339090,"url":"https://github.com/capsadmin/diffusers-a1111","last_synced_at":"2025-07-12T14:31:41.337Z","repository":{"id":181906363,"uuid":"667596548","full_name":"CapsAdmin/diffusers-a1111","owner":"CapsAdmin","description":"experimenting","archived":false,"fork":false,"pushed_at":"2023-07-26T10:20:43.000Z","size":88,"stargazers_count":9,"open_issues_count":0,"forks_count":2,"subscribers_count":2,"default_branch":"main","last_synced_at":"2024-05-02T01:06:40.933Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/CapsAdmin.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null}},"created_at":"2023-07-17T21:57:29.000Z","updated_at":"2024-01-10T02:53:37.000Z","dependencies_parsed_at":"2023-07-18T03:06:55.513Z","dependency_job_id":null,"html_url":"https://github.com/CapsAdmin/diffusers-a1111","commit_stats":null,"previous_names":["capsadmin/diffusers-a1111"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/CapsAdmin%2Fdiffusers-a1111","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/CapsAdmin%2Fdiffusers-a1111/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/CapsAdmin%2Fdiffusers-a1111/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/CapsAdmin%2Fdiffusers-a1111/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/CapsAdmin","download_url":"https://codeload.github.com/CapsAdmin/diffusers-a1111/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":225825236,"owners_count":17529905,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2024-11-22T00:42:20.450Z","updated_at":"2024-11-22T00:42:20.893Z","avatar_url":"https://github.com/CapsAdmin.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Setup\n\n    python3.10 -m venv .venv\n    python3.10 -m pip install --pre torch torchvision torchaudio --index-url https://download.pytorch.org/whl/nightly/rocm5.5\n    python3.10 -m pip install -r requirements.txt\n\n# Currently implemented\n\nThis implements a lot of the txt2img features from a1111 with diffusers. The goal is to be able to reproduce results from a1111 in diffusers\n\n- seed resize\n- variation seed\n- k-diffusion samplers\n- advanced sampler parameters\n- clip skip\n- prompt weighing\n- lora, lokr, ia3, hada and full\n\n# Not yet implemented\n\n- hypernetworks\n- face upsacling\n- tile\n- highres fix\n- img2img and inpaint\n- controlnet extension (some work)\n- AND, and other prompt features\n- batch size and seed logic\n\n```python\nfrom txt2img import txt2img\n\nimage = txt2img(\n    checkpoint = \"juggernaut_final\",\n\n    positive = \"\"\"(old man:1.3)  \u003clora:studioGhibliStyle_offset:1\u003e\"\"\",\n    negative = \"\"\"\"\"\",\n\n    steps = 20,\n    seed = 1337,\n\n    width = 512,\n    height = 512,\n\n    cfg_scale = 7.5,\n\n    subseed_strength = 0,\n    sub_seed = 1,\n    seed_resize_from_h = 0,\n    seed_resize_from_w = 0,\n\n    # Clip skip (ignore last layers of CLIP network; 1 ignores none, 2 ignores one layer) https://github.com/AUTOMATIC1111/stable-diffusion-webui/wiki/Features#clip-skip\n    CLIP_stop_at_last_layers = 1,\n\n    # Which algorithm is used to produce the image\n    sampler_name = \"UniPC\",\n\n    # Enable quantization in K samplers for sharper and cleaner results. This may change existing seeds. Requires restart to apply.\n    enable_quantization = True,\n\n    # Use old karras scheduler sigmas (0.1 to 10).\n    use_old_karras_scheduler_sigmas = False,\n\n    # Do not make DPM++ SDE deterministic across different batch sizes.\n    no_dpmpp_sde_batch_determinism = False,\n\n    # Do not fix prompt schedule for second order samplers.\n    dont_fix_second_order_samplers_schedule = False,\n\n    # Make K-diffusion samplers produce same images in a batch as when making a single image\n    k_diffusion_batch = True,\n\n    # Random number generator source. (changes seeds drastically; use CPU to produce the same picture across different videocard vendors)\n    randn_source = \"gpu\",\n\n    # sampler parameters from a1111\n\n    # these affect sample_euler, sample_heun and sample_dpm_2\n    s_churn = 0.0,\n    s_tmin = 0.0,\n    s_tmax = 0.0, # this doesn't actually exist in the a1111 ui, but it's in the code\n    s_noise = 1.0,\n\n    # scheduler type (lets you override the noise schedule for k-diffusion samplers; choosing Automatic disables the three parameters below)\n    k_sched_type = \"Automatic\",\n\n    # sigma min (0 = default (~0.03); minimum noise strength for k-diffusion noise scheduler)\n    sigma_min = 0,\n\n    # sigma max (0 = default (~14.6); maximum noise strength for k-diffusion noise schedule)\n    sigma_max = 0,\n\n    #  rho (0 = default (7 for karras, 1 for polyexponential); higher values result in a more steep noise schedule (decreases faster))\n    rho = 0.0,\n\n    # Eta noise seed delta (ENSD; does not improve anything, just produces different results for ancestral samplers - only useful for reproducing images)\n    eta_noise_seed_delta = 0,\n\n    # Eta for ancestral samplers, noise multiplier; applies to Euler a and other samplers that have a in them\n    eta_ancestral = 1.0,\n\n    # Eta for DDIM, noise multiplier; higher = more unperdictable results\n    eta_ddim = 0.0,\n\n    ddim_discretize = \"uniform\", # quad\n\n    # Always discard next-to-last sigma https://github.com/AUTOMATIC1111/stable-diffusion-webui/pull/6044\n    always_discard_next_to_last_sigma = False,\n\n    # UniPC variant\n    uni_pc_variant = \"bh1\", # bh2, vary_coef\n\n    # UniPC skip type\n    uni_pc_skip_type = \"time_uniform\", # time_quadratic logSNR\n\n    # UniPC order (must be \u003c sampling steps)\n    uni_pc_order = 3,\n\n    # UniPC lower order final\n    uni_pc_lower_order_final = True,\n)\n\nimage.save(\"../output/output.png\")\n```\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fcapsadmin%2Fdiffusers-a1111","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fcapsadmin%2Fdiffusers-a1111","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fcapsadmin%2Fdiffusers-a1111/lists"}