What is AuraFlow?
AuraFlow is a flow-based text2image model developed by fal.ai. It is released under the Apache-2.0 license and is characterized by being lightweight and easy to handle.
There are some movements to move from SDXL, and Pony V7, a representative example of this, is also introduced here.
Model Download
The latest version is AuraFlow v0.3.
📂ComfyUI/
└── 📂models/
└── 📂checkpoints/
└── aura_flow_0.3.safetensors
text2image
The basic construction is almost the same as SD1.5 / SDXL.
{
"id": "fdf97cd0-386c-47ac-af66-85d545ef1b28",
"revision": 0,
"last_node_id": 7,
"last_link_id": 9,
"nodes": [
{
"id": 1,
"type": "VAEDecode",
"pos": [
1216.6246298006072,
355.1087682795641
],
"size": [
210,
46
],
"flags": {},
"order": 5,
"mode": 0,
"inputs": [
{
"name": "samples",
"type": "LATENT",
"link": 1
},
{
"name": "vae",
"type": "VAE",
"link": 2
}
],
"outputs": [
{
"name": "IMAGE",
"type": "IMAGE",
"slot_index": 0,
"links": [
3
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.33",
"Node name for S&R": "VAEDecode"
}
},
{
"id": 2,
"type": "EmptyLatentImage",
"pos": [
589.7596615388886,
773.6887682795641
],
"size": [
244.81999999999994,
106
],
"flags": {},
"order": 0,
"mode": 0,
"inputs": [],
"outputs": [
{
"name": "LATENT",
"type": "LATENT",
"slot_index": 0,
"links": [
9
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.33",
"Node name for S&R": "EmptyLatentImage"
},
"widgets_values": [
1216,
832,
1
]
},
{
"id": 3,
"type": "SaveImage",
"pos": [
1458.6246298006072,
356.1087682795641
],
"size": [
473.65578531865367,
392.65694594515685
],
"flags": {},
"order": 6,
"mode": 0,
"inputs": [
{
"name": "images",
"type": "IMAGE",
"link": 3
}
],
"outputs": [],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.33"
},
"widgets_values": [
"ComfyUI"
]
},
{
"id": 4,
"type": "CLIPTextEncode",
"pos": [
422.6246298006073,
353.1087682795641
],
"size": [
411.95503173828126,
151.0030493164063
],
"flags": {},
"order": 2,
"mode": 0,
"inputs": [
{
"name": "clip",
"type": "CLIP",
"link": 4
}
],
"outputs": [
{
"name": "CONDITIONING",
"type": "CONDITIONING",
"slot_index": 0,
"links": [
7
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.33",
"Node name for S&R": "CLIPTextEncode"
},
"widgets_values": [
"A dog on a log with a frog in a bgog"
]
},
{
"id": 5,
"type": "CLIPTextEncode",
"pos": [
423.8216464021698,
559.4872533869859
],
"size": [
410.75801513671877,
158.82607910156253
],
"flags": {},
"order": 3,
"mode": 0,
"inputs": [
{
"name": "clip",
"type": "CLIP",
"link": 5
}
],
"outputs": [
{
"name": "CONDITIONING",
"type": "CONDITIONING",
"slot_index": 0,
"links": [
8
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.33",
"Node name for S&R": "CLIPTextEncode"
},
"widgets_values": [
"text, watermark"
]
},
{
"id": 6,
"type": "CheckpointLoaderSimple",
"pos": [
60.07098592382306,
468.50139788136767
],
"size": [
315,
98
],
"flags": {},
"order": 1,
"mode": 0,
"inputs": [],
"outputs": [
{
"name": "MODEL",
"type": "MODEL",
"slot_index": 0,
"links": [
6
]
},
{
"name": "CLIP",
"type": "CLIP",
"slot_index": 1,
"links": [
4,
5
]
},
{
"name": "VAE",
"type": "VAE",
"slot_index": 2,
"links": [
2
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.33",
"Node name for S&R": "CheckpointLoaderSimple"
},
"widgets_values": [
"AuraFlow\\aura_flow_0.3.safetensors"
],
"color": "#323",
"bgcolor": "#535"
},
{
"id": 7,
"type": "KSampler",
"pos": [
870.6246298006073,
353.1087682795641
],
"size": [
315,
262
],
"flags": {},
"order": 4,
"mode": 0,
"inputs": [
{
"name": "model",
"type": "MODEL",
"link": 6
},
{
"name": "positive",
"type": "CONDITIONING",
"link": 7
},
{
"name": "negative",
"type": "CONDITIONING",
"link": 8
},
{
"name": "latent_image",
"type": "LATENT",
"link": 9
}
],
"outputs": [
{
"name": "LATENT",
"type": "LATENT",
"slot_index": 0,
"links": [
1
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.33",
"Node name for S&R": "KSampler"
},
"widgets_values": [
1234,
"fixed",
20,
3.5,
"euler",
"sgm_uniform",
1
]
}
],
"links": [
[
1,
7,
0,
1,
0,
"LATENT"
],
[
2,
6,
2,
1,
1,
"VAE"
],
[
3,
1,
0,
3,
0,
"IMAGE"
],
[
4,
6,
1,
4,
0,
"CLIP"
],
[
5,
6,
1,
5,
0,
"CLIP"
],
[
6,
6,
0,
7,
0,
"MODEL"
],
[
7,
4,
0,
7,
1,
"CONDITIONING"
],
[
8,
5,
0,
7,
2,
"CONDITIONING"
],
[
9,
2,
0,
7,
3,
"LATENT"
]
],
"groups": [],
"config": {},
"extra": {
"ds": {
"scale": 0.8371166402407754,
"offset": [
39.92901407617694,
-253.10876827956412
]
},
"frontendVersion": "1.35.0",
"VHS_latentpreview": false,
"VHS_latentpreviewrate": 0,
"VHS_MetadataImage": true,
"VHS_KeepIntermediate": true
},
"version": 0.4
}
Pony V7 : AuraFlow-based Anime Model
It is an anime-oriented model based on AuraFlow, created as a successor to Pony Diffusion V6 XL.
Model Download
- diffusion_model
- text_encoder
- VAE
📂ComfyUI/
└── 📂models/
├── 📂diffusion_models/
│ └── pony-v7-base.safetensors
├── 📂text_encoders/
│ └── model.fp16.safetensors
└── 📂vae/
└── diffusion_pytorch_model.fp16.safetensors
text2image
{
"id": "96cd0bc1-1373-41e4-8208-b6c77cbcae42",
"revision": 0,
"last_node_id": 20,
"last_link_id": 23,
"nodes": [
{
"id": 13,
"type": "UNETLoader",
"pos": [
564.2086681019177,
35.10858985725014
],
"size": [
273.6363636363635,
82
],
"flags": {},
"order": 0,
"mode": 0,
"inputs": [],
"outputs": [
{
"name": "MODEL",
"type": "MODEL",
"links": [
13
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.66",
"Node name for S&R": "UNETLoader"
},
"widgets_values": [
"AuraFlow\\pony-v7-base.safetensors",
"default"
],
"color": "#323",
"bgcolor": "#535"
},
{
"id": 8,
"type": "VAEDecode",
"pos": [
1216.974250392735,
188.47933884297524
],
"size": [
150.10743801652893,
46
],
"flags": {},
"order": 9,
"mode": 0,
"inputs": [
{
"name": "samples",
"type": "LATENT",
"link": 7
},
{
"name": "vae",
"type": "VAE",
"link": 12
}
],
"outputs": [
{
"name": "IMAGE",
"type": "IMAGE",
"slot_index": 0,
"links": [
16
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.66",
"Node name for S&R": "VAEDecode"
},
"widgets_values": []
},
{
"id": 14,
"type": "CLIPLoader",
"pos": [
-158.19999677617656,
302.78019698791024
],
"size": [
276.2236186052865,
106
],
"flags": {},
"order": 1,
"mode": 0,
"inputs": [],
"outputs": [
{
"name": "CLIP",
"type": "CLIP",
"links": [
21
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.66",
"Node name for S&R": "CLIPLoader"
},
"widgets_values": [
"model.fp16.safetensors",
"qwen_image",
"default"
],
"color": "#432",
"bgcolor": "#653"
},
{
"id": 16,
"type": "T5TokenizerOptions",
"pos": [
142.0583843829231,
305.356336880927
],
"size": [
210,
82
],
"flags": {},
"order": 5,
"mode": 0,
"inputs": [
{
"name": "clip",
"type": "CLIP",
"link": 21
}
],
"outputs": [
{
"name": "CLIP",
"type": "CLIP",
"links": [
17,
18
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.66",
"Node name for S&R": "T5TokenizerOptions"
},
"widgets_values": [
768,
768
],
"color": "#223",
"bgcolor": "#335"
},
{
"id": 11,
"type": "EmptyLatentImage",
"pos": [
606.0268499200995,
613.2091909090913
],
"size": [
231.81818181818176,
106
],
"flags": {},
"order": 2,
"mode": 0,
"inputs": [],
"outputs": [
{
"name": "LATENT",
"type": "LATENT",
"links": [
23
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.66",
"Node name for S&R": "EmptyLatentImage"
},
"widgets_values": [
1280,
768,
1
]
},
{
"id": 6,
"type": "CLIPTextEncode",
"pos": [
415,
186
],
"size": [
422.84503173828125,
164.31304931640625
],
"flags": {},
"order": 6,
"mode": 0,
"inputs": [
{
"name": "clip",
"type": "CLIP",
"link": 17
}
],
"outputs": [
{
"name": "CONDITIONING",
"type": "CONDITIONING",
"slot_index": 0,
"links": [
4
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.66",
"Node name for S&R": "CLIPTextEncode"
},
"widgets_values": [
"style_cluster_1610, score_9, rating_safe,\ncool anime style, MAPPA-like line art, clean thin lines, subtle cel shading,\nrefined realistic lighting, cinematic composition, shallow depth of field, realistic photographic background, filmic color tone, slightly desaturated palette,\na drenched office woman in a navy tailored business suit (single-breasted jacket, white blouse),\nan inflatable swim ring partially visible around her torso, upper-body close-up, hands resting on the rim,\nwet hair clinging, soaked fabric contouring, water droplets, tired blank upward gaze,\ncentered composition, tight framing,\ngentle ripples and sunlight reflections on blue water, serene yet melancholic mood\n"
]
},
{
"id": 7,
"type": "CLIPTextEncode",
"pos": [
414.05164330852915,
406.27272727272765
],
"size": [
423.7933884297521,
147.40702479338836
],
"flags": {},
"order": 7,
"mode": 0,
"inputs": [
{
"name": "clip",
"type": "CLIP",
"link": 18
}
],
"outputs": [
{
"name": "CONDITIONING",
"type": "CONDITIONING",
"slot_index": 0,
"links": [
6
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.66",
"Node name for S&R": "CLIPTextEncode"
},
"widgets_values": [
"full body, lower body, legs, distant view, wide shot, small figure, empty swim ring, text, logo, watermark\n"
]
},
{
"id": 20,
"type": "MarkdownNote",
"pos": [
191.83015626879293,
-102.29636882043582
],
"size": [
339.7520661157024,
219.40495867768595
],
"flags": {},
"order": 3,
"mode": 0,
"inputs": [],
"outputs": [],
"properties": {},
"widgets_values": [
"## models\n- [pony-v7-base.safetensors](https://huggingface.co/purplesmartai/pony-v7-base/blob/main/safetensor/pony-v7-base.safetensors)\n- [text_encoder/model.fp16.safetensors](https://huggingface.co/purplesmartai/pony-v7-base/blob/main/text_encoder/model.fp16.safetensors)\n- [vae/diffusion_pytorch_model.fp16.safetensors](https://huggingface.co/purplesmartai/pony-v7-base/blob/main/vae/diffusion_pytorch_model.fp16.safetensors)\n\n```\n📂ComfyUI/\n└── 📂models/\n ├── 📂diffusion_models/\n │ └── pony-v7-base.safetensors\n ├── 📂text_encoders/\n │ └── model.fp16.safetensors\n └── 📂vae/\n └── diffusion_pytorch_model.fp16.safetensors\n```"
],
"color": "#323",
"bgcolor": "#535"
},
{
"id": 12,
"type": "VAELoader",
"pos": [
883.9505393893851,
62.783748657878505
],
"size": [
289.91257427771325,
58
],
"flags": {},
"order": 4,
"mode": 0,
"inputs": [],
"outputs": [
{
"name": "VAE",
"type": "VAE",
"links": [
12
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.66",
"Node name for S&R": "VAELoader"
},
"widgets_values": [
"diffusion_pytorch_model.fp16.safetensors"
],
"color": "#322",
"bgcolor": "#533"
},
{
"id": 15,
"type": "SaveImage",
"pos": [
1395.162538501471,
186.18476218154478
],
"size": [
547.5454545454547,
467.81818181818176
],
"flags": {},
"order": 10,
"mode": 0,
"inputs": [
{
"name": "images",
"type": "IMAGE",
"link": 16
}
],
"outputs": [],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.66"
},
"widgets_values": [
"ComfyUI"
]
},
{
"id": 3,
"type": "KSampler",
"pos": [
884.0495867768592,
188.47933884297524
],
"size": [
291.0945290622226,
262
],
"flags": {},
"order": 8,
"mode": 0,
"inputs": [
{
"name": "model",
"type": "MODEL",
"link": 13
},
{
"name": "positive",
"type": "CONDITIONING",
"link": 4
},
{
"name": "negative",
"type": "CONDITIONING",
"link": 6
},
{
"name": "latent_image",
"type": "LATENT",
"link": 23
}
],
"outputs": [
{
"name": "LATENT",
"type": "LATENT",
"slot_index": 0,
"links": [
7
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.66",
"Node name for S&R": "KSampler"
},
"widgets_values": [
225489861693909,
"fixed",
30,
1.5,
"euler_cfg_pp",
"sgm_uniform",
1
]
}
],
"links": [
[
4,
6,
0,
3,
1,
"CONDITIONING"
],
[
6,
7,
0,
3,
2,
"CONDITIONING"
],
[
7,
3,
0,
8,
0,
"LATENT"
],
[
12,
12,
0,
8,
1,
"VAE"
],
[
13,
13,
0,
3,
0,
"MODEL"
],
[
16,
8,
0,
15,
0,
"IMAGE"
],
[
17,
16,
0,
6,
0,
"CLIP"
],
[
18,
16,
0,
7,
0,
"CLIP"
],
[
21,
14,
0,
16,
0,
"CLIP"
],
[
23,
11,
0,
3,
3,
"LATENT"
]
],
"groups": [],
"config": {},
"extra": {
"ds": {
"scale": 0.751314800901578,
"offset": [
258.1999967761766,
202.29636882043582
]
},
"frontendVersion": "1.35.0",
"VHS_latentpreview": false,
"VHS_latentpreviewrate": 0,
"VHS_MetadataImage": true,
"VHS_KeepIntermediate": true
},
"version": 0.4
}
- 🟦
T5TokenizerOptions- A node that fills with padding if the number of tokens is below the set value. There is no big difference whether you include it or not.
- The official workflow was
euler_normal, but since the lines tended to be chaotic, CFG++ (Improved CFG Guidance) is used here.- It adjusts conventional CFG smoothly, making lines easier to stabilize.