From a5e48c66aacb0a7be9b73fa3b84df9a6829578a1 Mon Sep 17 00:00:00 2001 From: leejet Date: Sun, 25 Oct 2026 19:37:41 +0800 Subject: [PATCH] chore: reduce large recipe defaults to 1024x1024 except PiD --- examples/server/recipes/defaults.md | 25 +++++++++++-------- examples/server/recipes/flux1-fill.json | 4 +-- examples/server/recipes/hidream-o1-dev.json | 4 +-- examples/server/recipes/krea2-turbo-edit.json | 4 +-- examples/server/recipes/krea2-turbo.json | 4 +-- examples/server/recipes/lens-turbo.json | 4 +-- examples/server/recipes/lens.json | 4 +-- examples/server/recipes/ming-image.json | 4 +-- .../server/recipes/qwen-image-21-turbo.json | 4 +-- examples/server/recipes/qwen-image-21.json | 4 +-- examples/server/recipes/qwen-image.json | 4 +-- examples/server/recipes/sensenova-u15.json | 4 +-- 12 files changed, 36 insertions(+), 33 deletions(-) diff --git a/examples/server/recipes/defaults.md b/examples/server/recipes/defaults.md index 2be4dfbf..e20a2cd7 100644 --- a/examples/server/recipes/defaults.md +++ b/examples/server/recipes/defaults.md @@ -5,6 +5,9 @@ Reviewed against model-publisher cards and reference implementations on 2026-10- These are starting values, not limits on supported resolutions or generation settings. Built-in step defaults above 20 are reduced to 20; the cited sources may recommend more steps. Users can still choose a higher step count. +Except for PiD recipes, built-in dimensions above 1,048,576 pixels (1024 x 1024) are reduced to 1024 x 1024, even when the cited source uses a larger size. +PiD recipes retain their 2048 x 2048 defaults. +Users can still choose a higher resolution. When several aspect ratios or ranges are published, the selected example or square bucket is noted below. A dash means no recipe override: the available primary sources did not establish a fixed value, or the size depends on the input image. Ordinary DiT recipes explicitly use Euler; PiD retains its LCM sampling path. @@ -27,7 +30,7 @@ For community edit adapters and other optional extensions, these are the base-mo | `flux1-canny` | 1 | 20 | 1024 × 1024 | [Source](https://huggingface.co/black-forest-labs/FLUX.1-Canny-dev) | | `flux1-depth` | 1 | 20 | 1024 × 1024 | [Source](https://huggingface.co/black-forest-labs/FLUX.1-Depth-dev) | | `flux1-dev` | 1 | 20 | 1024 × 1024 | [Source](https://huggingface.co/black-forest-labs/FLUX.1-dev) | -| `flux1-fill` | 1 | 20 | 1232 × 1632 | [Source](https://huggingface.co/black-forest-labs/FLUX.1-Fill-dev) | +| `flux1-fill` | 1 | 20 | 1024 × 1024 | [Source](https://huggingface.co/black-forest-labs/FLUX.1-Fill-dev) | | `flux1-kontext` | 1 | 20 | — | [Source](https://raw.githubusercontent.com/huggingface/diffusers/main/src/diffusers/pipelines/flux/pipeline_flux_kontext.py) Output size normally follows the reference image; no fixed size is imposed by this recipe. | | `flux1-pulid` | 1 | 20 | 1024 × 1024 | [Source](https://huggingface.co/black-forest-labs/FLUX.1-dev) FLUX.1-dev defaults; identity conditioning is provided by PuLID. | | `flux1-schnell` | 1 | 4 | 1024 × 1024 | [Source](https://huggingface.co/black-forest-labs/FLUX.1-schnell) | @@ -38,7 +41,7 @@ For community edit adapters and other optional extensions, these are the base-mo | `flux2-klein-base-9b` | 4 | 20 | 1024 × 1024 | [Source](https://huggingface.co/black-forest-labs/FLUX.2-klein-base-9B) | | `flux3-klein` | 1 | 16 | 1024 × 1024 | [Source](https://raw.githubusercontent.com/black-forest-labs/flux-3-image-klein/main/README.md) | | `flux3-klein-base` | 4 | 20 | 1024 × 1024 | [Source](https://raw.githubusercontent.com/black-forest-labs/flux-3-image-klein/main/src/flux3/pipeline.py) | -| `hidream-o1-dev` | 1 | 20 | 2048 × 2048 | [Source](https://huggingface.co/HiDream-ai/HiDream-O1-Image-Dev) | +| `hidream-o1-dev` | 1 | 20 | 1024 × 1024 | [Source](https://huggingface.co/HiDream-ai/HiDream-O1-Image-Dev) | | `hunyuan-video-15-i2v` | 6 | 20 | 1280 × 720 | [Source](https://huggingface.co/tencent/HunyuanVideo-1.5) 720p base I2V defaults; CFG/step-distilled and 480p checkpoints use different settings. | | `hunyuan-video-15-t2v` | 6 | 20 | 1280 × 720 | [Source](https://huggingface.co/tencent/HunyuanVideo-1.5) 720p base T2V defaults; CFG-distilled / 480p checkpoints need their published settings. | | `ideogram4` | 7 | 20 | 1024 × 1024 | [Source](https://raw.githubusercontent.com/ideogram-oss/ideogram4/main/docs/inference.md) Constant-guidance pipeline defaults; this recipe does not reproduce the CLI’s per-step guidance schedule. | @@ -46,10 +49,10 @@ For community edit adapters and other optional extensions, these are the base-mo | `iris-3b` | 3 | 20 | 1024 × 1024 | [Source](https://huggingface.co/speridlabs/iris-3b) | | `krea2-raw` | 3.5 | 20 | 1024 × 1024 | [Source](https://huggingface.co/krea/Krea-2-Raw) Base model defaults; edit requires a matching community LoRA with its own settings. | | `krea2-raw-edit` | 3.5 | 20 | 1024 × 1024 | [Source](https://huggingface.co/krea/Krea-2-Raw) Base model defaults; edit requires a matching community LoRA with its own settings. | -| `krea2-turbo` | 1 | 8 | 2048 × 2048 | [Source](https://huggingface.co/krea/Krea-2-Turbo) Official CLI example; edit requires a matching community LoRA. | -| `krea2-turbo-edit` | 1 | 8 | 2048 × 2048 | [Source](https://huggingface.co/krea/Krea-2-Turbo) Official CLI example; edit requires a matching community LoRA. | -| `lens` | 5 | 20 | 1440 × 1440 | [Source](https://huggingface.co/microsoft/Lens) Official model card: RL model, default base_resolution 1440. | -| `lens-turbo` | 1 | 4 | 1440 × 1440 | [Source](https://huggingface.co/microsoft/Lens-Turbo) Official model card: 4-step distilled model, default base_resolution 1440. | +| `krea2-turbo` | 1 | 8 | 1024 × 1024 | [Source](https://huggingface.co/krea/Krea-2-Turbo) Official CLI example; edit requires a matching community LoRA. | +| `krea2-turbo-edit` | 1 | 8 | 1024 × 1024 | [Source](https://huggingface.co/krea/Krea-2-Turbo) Official CLI example; edit requires a matching community LoRA. | +| `lens` | 5 | 20 | 1024 × 1024 | [Source](https://huggingface.co/microsoft/Lens) Official model card: RL model, default base_resolution 1440. | +| `lens-turbo` | 1 | 4 | 1024 × 1024 | [Source](https://huggingface.co/microsoft/Lens-Turbo) Official model card: 4-step distilled model, default base_resolution 1440. | | `lingbot-video-dense` | 3 | 20 | 832 × 480 | [Source](https://huggingface.co/robbyant/lingbot-video-dense-1.3b) Official first-stage inference example; optional refiner is separate. | | `lingbot-video-moe` | 3 | 20 | 832 × 480 | [Source](https://huggingface.co/robbyant/lingbot-video-moe-30b-a3b) Official inference example. | | `llada-image` | 5 | 20 | 1024 × 1024 | [Source](https://huggingface.co/inclusionAI/LLaDA-Image) | @@ -62,7 +65,7 @@ For community edit adapters and other optional extensions, these are the base-mo | `mage-flow` | 5 | 20 | 1024 × 1024 | [Source](https://huggingface.co/microsoft/Mage-Flow) | | `mage-flow-edit` | 5 | 20 | 1024 × 1024 | [Source](https://huggingface.co/microsoft/Mage-Flow) | | `mage-flow-turbo` | 1 | 4 | 1024 × 1024 | [Source](https://huggingface.co/microsoft/Mage-Flow) | -| `ming-image` | 1 | 12 | 2048 × 2048 | [Source](https://huggingface.co/inclusionAI/Ming-Image-0.1-Design) | +| `ming-image` | 1 | 12 | 1024 × 1024 | [Source](https://huggingface.co/inclusionAI/Ming-Image-0.1-Design) | | `minimax-h3-fl2va` | 1 | — | 1344 × 768 | [Source](https://huggingface.co/docs/diffusers/main/en/api/pipelines/minimax_h3) Guidance-distilled; official 768p canvas. No fixed step count is prescribed here; the upstream API counts sigma grid points rather than sd.cpp denoising steps. | | `minimax-h3-ref2va` | 1 | — | 1344 × 768 | [Source](https://huggingface.co/docs/diffusers/main/en/api/pipelines/minimax_h3) Guidance-distilled; official 768p canvas. No fixed step count is prescribed here; the upstream API counts sigma grid points rather than sd.cpp denoising steps. | | `minit2i` | 2.5 | 20 | 512 × 512 | [Source](https://huggingface.co/MiniT2I/MiniT2I/raw/main/pipeline.py) B/16 example and native model size; L/16 uses CFG 6. | @@ -81,9 +84,9 @@ For community edit adapters and other optional extensions, these are the base-mo | `pixart-alpha-1024` | 4.5 | 20 | 1024 × 1024 | [Source](https://raw.githubusercontent.com/huggingface/diffusers/main/src/diffusers/pipelines/pixart_alpha/pipeline_pixart_alpha.py) | | `pixart-alpha-512` | 4.5 | 20 | 512 × 512 | [Source](https://raw.githubusercontent.com/huggingface/diffusers/main/src/diffusers/pipelines/pixart_alpha/pipeline_pixart_alpha.py) | | `pixart-sigma` | 4.5 | 20 | 1024 × 1024 | [Source](https://raw.githubusercontent.com/huggingface/diffusers/main/src/diffusers/pipelines/pixart_alpha/pipeline_pixart_sigma.py) | -| `qwen-image` | 4 | 20 | 1328 × 1328 | [Source](https://huggingface.co/Qwen/Qwen-Image) Official square resolution bucket; true_cfg_scale maps to cfg_scale. | -| `qwen-image-21` | 1 | 20 | 2048 × 2048 | [Source](https://raw.githubusercontent.com/QwenLM/Qwen-Image-2.1/main/README.md) Original checkpoint defaults; select the matching TAEQI2.1 to replace its VAE. | -| `qwen-image-21-turbo` | 1 | 8 | 2048 × 2048 | [Model card](https://huggingface.co/Qwen/Qwen-Image-2.1-Turbo), [fixed sigmas](https://huggingface.co/Qwen/Qwen-Image-2.1-Turbo/blob/main/model_index.json). Standalone Turbo weights; the fixed 8-step schedule is applied on model load. | +| `qwen-image` | 4 | 20 | 1024 × 1024 | [Source](https://huggingface.co/Qwen/Qwen-Image) The source uses a larger square resolution bucket; true_cfg_scale maps to cfg_scale. | +| `qwen-image-21` | 1 | 20 | 1024 × 1024 | [Source](https://raw.githubusercontent.com/QwenLM/Qwen-Image-2.1/main/README.md) Original checkpoint sampling defaults; select the matching TAEQI2.1 to replace its VAE. | +| `qwen-image-21-turbo` | 1 | 8 | 1024 × 1024 | [Model card](https://huggingface.co/Qwen/Qwen-Image-2.1-Turbo), [fixed sigmas](https://huggingface.co/Qwen/Qwen-Image-2.1-Turbo/blob/main/model_index.json). Standalone Turbo weights; the fixed 8-step schedule is applied on model load. | | `qwen-image-edit` | 4 | 20 | — | [Source](https://huggingface.co/Qwen/Qwen-Image-Edit) Official editing example; output dimensions follow reference-image aspect ratio. | | `qwen-image-edit-2509` | 4 | 20 | — | [Source](https://huggingface.co/Qwen/Qwen-Image-Edit-2509) Official editing example; output dimensions follow reference-image aspect ratio. | | `qwen-image-edit-2511` | 4 | 20 | — | [Source](https://huggingface.co/Qwen/Qwen-Image-Edit-2511) Official editing example; output dimensions follow reference-image aspect ratio. | @@ -107,7 +110,7 @@ For community edit adapters and other optional extensions, these are the base-mo | `sdxl-turbo` | 1 | 1 | 512 × 512 | [Source](https://huggingface.co/stabilityai/sdxl-turbo) | | `sdxs` | 1 | 1 | 512 × 512 | [Source](https://huggingface.co/IDKiro/sdxs-512-dreamshaper) | | `sefi-image` | 4 | 20 | — | [Source](https://huggingface.co/SeFi-Image/SeFi-Image-5B-Base) Base/RL family defaults; Turbo checkpoints require 4 steps / CFG 1. | -| `sensenova-u15` | 4 | 20 | 2048 × 2048 | [Source](https://raw.githubusercontent.com/OpenSenseNova/SenseNova-U1/main/examples/README.md) | +| `sensenova-u15` | 4 | 20 | 1024 × 1024 | [Source](https://raw.githubusercontent.com/OpenSenseNova/SenseNova-U1/main/examples/README.md) | | `wan21-flf2v` | 5 | 20 | 1280 × 720 | [Source](https://raw.githubusercontent.com/Wan-Video/Wan2.1/main/generate.py) | | `wan21-i2v` | 5 | 20 | 832 × 480 | [Source](https://raw.githubusercontent.com/Wan-Video/Wan2.1/main/generate.py) 480p base defaults; also accepts compatible SkyReels V2 and single Wan 2.2 I2V experts. | | `wan21-t2v` | 5 | 20 | 1280 × 720 | [Source](https://raw.githubusercontent.com/Wan-Video/Wan2.1/main/generate.py) 14B base defaults; single Wan 2.2 experts and acceleration LoRAs need matching settings. | diff --git a/examples/server/recipes/flux1-fill.json b/examples/server/recipes/flux1-fill.json index 5e6485ad..fe27a48e 100644 --- a/examples/server/recipes/flux1-fill.json +++ b/examples/server/recipes/flux1-fill.json @@ -39,8 +39,8 @@ "scheduler": "flux", "cfg_scale": 1, "steps": 20, - "width": 1232, - "height": 1632 + "width": 1024, + "height": 1024 }, "features": [ "clip_skip", diff --git a/examples/server/recipes/hidream-o1-dev.json b/examples/server/recipes/hidream-o1-dev.json index 8e95f5ce..cf1c1790 100644 --- a/examples/server/recipes/hidream-o1-dev.json +++ b/examples/server/recipes/hidream-o1-dev.json @@ -13,8 +13,8 @@ "defaults": { "cfg_scale": 1, "steps": 20, - "width": 2048, - "height": 2048, + "width": 1024, + "height": 1024, "sampling_method": "euler" }, "features": [ diff --git a/examples/server/recipes/krea2-turbo-edit.json b/examples/server/recipes/krea2-turbo-edit.json index fa248c08..e8c521d3 100644 --- a/examples/server/recipes/krea2-turbo-edit.json +++ b/examples/server/recipes/krea2-turbo-edit.json @@ -39,8 +39,8 @@ "sampling_method": "euler", "cfg_scale": 1, "steps": 8, - "width": 2048, - "height": 2048 + "width": 1024, + "height": 1024 }, "features": [ "flow_shift", diff --git a/examples/server/recipes/krea2-turbo.json b/examples/server/recipes/krea2-turbo.json index 23dad12f..f19513f7 100644 --- a/examples/server/recipes/krea2-turbo.json +++ b/examples/server/recipes/krea2-turbo.json @@ -32,8 +32,8 @@ "defaults": { "cfg_scale": 1, "steps": 8, - "width": 2048, - "height": 2048, + "width": 1024, + "height": 1024, "sampling_method": "euler" }, "features": [ diff --git a/examples/server/recipes/lens-turbo.json b/examples/server/recipes/lens-turbo.json index e9282eaf..2b87bb64 100644 --- a/examples/server/recipes/lens-turbo.json +++ b/examples/server/recipes/lens-turbo.json @@ -36,8 +36,8 @@ "defaults": { "cfg_scale": 1, "steps": 4, - "width": 1440, - "height": 1440, + "width": 1024, + "height": 1024, "sampling_method": "euler" }, "features": [ diff --git a/examples/server/recipes/lens.json b/examples/server/recipes/lens.json index 0a895422..34165ad5 100644 --- a/examples/server/recipes/lens.json +++ b/examples/server/recipes/lens.json @@ -36,8 +36,8 @@ "defaults": { "cfg_scale": 5, "steps": 20, - "width": 1440, - "height": 1440, + "width": 1024, + "height": 1024, "sampling_method": "euler" }, "features": [ diff --git a/examples/server/recipes/ming-image.json b/examples/server/recipes/ming-image.json index 6c368e8e..465a690e 100644 --- a/examples/server/recipes/ming-image.json +++ b/examples/server/recipes/ming-image.json @@ -36,8 +36,8 @@ "defaults": { "cfg_scale": 1, "steps": 12, - "width": 2048, - "height": 2048, + "width": 1024, + "height": 1024, "sampling_method": "euler" }, "features": [ diff --git a/examples/server/recipes/qwen-image-21-turbo.json b/examples/server/recipes/qwen-image-21-turbo.json index 2ef9fa18..99da90d7 100644 --- a/examples/server/recipes/qwen-image-21-turbo.json +++ b/examples/server/recipes/qwen-image-21-turbo.json @@ -40,8 +40,8 @@ "sampling_method": "euler", "cfg_scale": 1, "steps": 8, - "width": 2048, - "height": 2048, + "width": 1024, + "height": 1024, "custom_sigmas": [ 1.0, 0.978453, diff --git a/examples/server/recipes/qwen-image-21.json b/examples/server/recipes/qwen-image-21.json index 4b6c67b8..fd487e58 100644 --- a/examples/server/recipes/qwen-image-21.json +++ b/examples/server/recipes/qwen-image-21.json @@ -40,8 +40,8 @@ "sampling_method": "euler", "cfg_scale": 1, "steps": 20, - "width": 2048, - "height": 2048 + "width": 1024, + "height": 1024 }, "features": [ "flow_shift", diff --git a/examples/server/recipes/qwen-image.json b/examples/server/recipes/qwen-image.json index 417a621a..468f8f26 100644 --- a/examples/server/recipes/qwen-image.json +++ b/examples/server/recipes/qwen-image.json @@ -35,8 +35,8 @@ "flow_shift": 3.0, "cfg_scale": 4, "steps": 20, - "width": 1328, - "height": 1328 + "width": 1024, + "height": 1024 }, "features": [ "flow_shift" diff --git a/examples/server/recipes/sensenova-u15.json b/examples/server/recipes/sensenova-u15.json index afd20466..f2df865e 100644 --- a/examples/server/recipes/sensenova-u15.json +++ b/examples/server/recipes/sensenova-u15.json @@ -14,8 +14,8 @@ "flow_shift": 3.0, "cfg_scale": 4, "steps": 20, - "width": 2048, - "height": 2048, + "width": 1024, + "height": 1024, "sampling_method": "euler" }, "features": [