aliyun-wan-image

Name: aliyun-wan-image
Author: cinience

62installs34forks

How to install aliyun-wan-image?

cinience/alicloud-skills/aliyun-wan-image

$npx -y skills add cinience/alicloud-skills --skill aliyun-wan-image

Installs into the current project.

Summary

Use when generating or editing images with DashScope Wan 2.7 image models (wan2.7-image, wan2.7-image-pro). Use when implementing text-to-image, image editing, interactive editing with bounding boxes, sequential group image generation, or color palette control via the multimodal-

Files · 1

View on GitHub

SKILL.md

1	# Wan 2.7 Image Generation & Editing
2
3	## Validation
4
5	```bash
6	mkdir -p output/aliyun-wan-image
7	python -m py_compile skills/ai/image/aliyun-wan-image/scripts/generate_image.py && echo "py_compile_ok" > output/aliyun-wan-image/validate.txt
8	```
9
10	Pass criteria: command exits 0 and `output/aliyun-wan-image/validate.txt` is generated.
11
12	## Output And Evidence
13
14	- Write generated image URLs, prompts, and metadata to `output/aliyun-wan-image/`.
15	- Keep at least one sample JSON response per run.
16
17	## Prerequisites
18
19	- Install SDK (recommended in a venv):
20
21	```bash
22	python3 -m venv .venv
23	. .venv/bin/activate
24	python -m pip install dashscope
25	```
26	- Set `DASHSCOPE_API_KEY` in your environment, or add `dashscope_api_key` to `~/.alibabacloud/credentials`.
27
28	## Critical model names
29
30	- `wan2.7-image-pro` — professional version, supports 4K output
31	- `wan2.7-image` — faster generation, up to 2K
32
33	## Capabilities
34
35	\| Capability \| Description \|
36	\|---\|---\|
37	\| Text-to-image \| Generate images from text prompts \|
38	\| Image editing \| Edit images with text instructions (1-9 input images) \|
39	\| Interactive editing \| Edit specific regions via bounding boxes (`bbox_list`) \|
40	\| Group generation \| Generate consistent multi-image sequences (`enable_sequential=true`, up to 12 images) \|
41	\| Color palette \| Control color theme with custom hex+ratio palette (3-10 colors) \|
42	\| Thinking mode \| Enhanced reasoning for better quality (text-to-image only) \|
43
44	## API endpoint
45
46	Sync (recommended):
47	```
48	POST https://dashscope.aliyuncs.com/api/v1/services/aigc/multimodal-generation/generation
49	```
50
51	Async (for long tasks):
52	```
53	POST https://dashscope.aliyuncs.com/api/v1/services/aigc/image-generation/generation
54	Header: X-DashScope-Async: enable
55	```
56
57	## Normalized interface (image.generate)
58
59	### Request
60	- `prompt` (string, required) — up to 5000 characters
61	- `size` (string, optional) — `1K`, `2K` (default), `4K` (pro only), or `WxH` pixel values
62	- `n` (int, optional) — number of images, 1-4 (default 4), or 1-12 with `enable_sequential`
63	- `seed` (int, optional) — range [0, 2147483647]
64	- `reference_image` (string/array, optional) — URL or base64, up to 9 images
65	- `enable_sequential` (bool, optional) — group image generation mode
66	- `thinking_mode` (bool, optional, default true) — enhanced reasoning (text-to-image only)
67	- `bbox_list` (array, optional) — bounding boxes for interactive editing
68	- `color_palette` (array, optional) — custom color theme (3-10 colors with hex+ratio)
69	- `watermark` (bool, optional, default false)
70
71	### Response
72	- `image_url` (string) — PNG, valid for 24 hours
73	- `image_count` (int)
74	- `size` (string) — actual output resolution
75	- `seed` (int)
76
77	## Quick start (Python + DashScope SDK)
78
79	```python
80	import os
81	from dashscope.aigc.image_generation import ImageGeneration
82
83	def generate_image(req: dict) -> dict:
84	messages = [
85	{
86	"role": "user",
87	"content": [{"text": req["prompt"]}],
88	}
89	]
90
91	# Add reference images if provided
92	ref_images = req.get("reference_images") or []
93	if req.get("reference_image"):
94	ref_images = [req["reference_image"]] + ref_images
95	for img in ref_images:
96	messages[0]["content"].append({"image": img})
97
98	params = {
99	"model": req.get("model", "wan2.7-image"),
100	"messages": messages,
101	"size": req.get("size", "2K"),
102	"n": req.get("n", 1),
103	"api_key": os.getenv("DASHSCOPE_API_KEY"),
104	"seed": req.get("seed"),
105	"watermark": req.get("watermark", False),
106	}
107
108	if req.get("enable_sequential"):
109	params["enable_sequential"] = True
110	if req.get("thinking_mode") is not None:
111	params["thinking_mode"] = req["thinking_mode"]
112	if req.get("bbox_list"):
113	params["bbox_list"] = req["bbox_list"]
114	if req.get("color_palette"):
115	params["color_palette"] = req["color_palette"]
116
117	response = ImageGeneration.call(**params)
118
119	content = response.output["choices"][0]["message"]["content"]
120	images = [item["image"] for item in content if isinstance(item, dict) and item.get("image")]
121
122	return {
123	"image_urls": images,
124	"image_count": response.usage.get("image_count"),
125	"size": response.usage.get("size"),
126	}
127	```
128
129	## Size reference
130
131	\| Model \| Supported sizes \| Default \|
132	\|---\|---\|---\|
133	\| wan2.7-image-pro \| 1K, 2K, 4K (text-to-image only), or [768, 4096] px \| 2K \|
134	\| wan2.7-image \| 1K, 2K, or [768, 2048] px \| 2K \|
135
136	## Error handling
137
138	\| Error \| Likely cause \| Action \|
139	\|---\|---\|---\|
140	\| 401/403 \| Missing or invalid `DASHSCOPE_API_KEY` \| Check env var or credentials file. \|
141	\| 400 `InvalidParameter` \| Unsupported size, bad n value, or missing required image \| Valida

People who liked this also viewed

6 picks

Type

aliyun-wan-image

62installs34forks

View on GitHub

How to install aliyun-wan-image?

cinience/alicloud-skills/aliyun-wan-image

$npx -y skills add cinience/alicloud-skills --skill aliyun-wan-image

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Wan 2.7 Image Generation & Editing
2
3	## Validation
4
5	```bash
6	mkdir -p output/aliyun-wan-image
7	python -m py_compile skills/ai/image/aliyun-wan-image/scripts/generate_image.py && echo "py_compile_ok" > output/aliyun-wan-image/validate.txt
8	```
9
10	Pass criteria: command exits 0 and `output/aliyun-wan-image/validate.txt` is generated.
11
12	## Output And Evidence
13
14	- Write generated image URLs, prompts, and metadata to `output/aliyun-wan-image/`.
15	- Keep at least one sample JSON response per run.
16
17	## Prerequisites
18
19	- Install SDK (recommended in a venv):
20
21	```bash
22	python3 -m venv .venv
23	. .venv/bin/activate
24	python -m pip install dashscope
25	```
26	- Set `DASHSCOPE_API_KEY` in your environment, or add `dashscope_api_key` to `~/.alibabacloud/credentials`.
27
28	## Critical model names
29
30	- `wan2.7-image-pro` — professional version, supports 4K output
31	- `wan2.7-image` — faster generation, up to 2K
32
33	## Capabilities
34
35	\| Capability \| Description \|
36	\|---\|---\|
37	\| Text-to-image \| Generate images from text prompts \|
38	\| Image editing \| Edit images with text instructions (1-9 input images) \|
39	\| Interactive editing \| Edit specific regions via bounding boxes (`bbox_list`) \|
40	\| Group generation \| Generate consistent multi-image sequences (`enable_sequential=true`, up to 12 images) \|
41	\| Color palette \| Control color theme with custom hex+ratio palette (3-10 colors) \|
42	\| Thinking mode \| Enhanced reasoning for better quality (text-to-image only) \|
43
44	## API endpoint
45
46	Sync (recommended):
47	```
48	POST https://dashscope.aliyuncs.com/api/v1/services/aigc/multimodal-generation/generation
49	```
50
51	Async (for long tasks):
52	```
53	POST https://dashscope.aliyuncs.com/api/v1/services/aigc/image-generation/generation
54	Header: X-DashScope-Async: enable
55	```
56
57	## Normalized interface (image.generate)
58
59	### Request
60	- `prompt` (string, required) — up to 5000 characters
61	- `size` (string, optional) — `1K`, `2K` (default), `4K` (pro only), or `WxH` pixel values
62	- `n` (int, optional) — number of images, 1-4 (default 4), or 1-12 with `enable_sequential`
63	- `seed` (int, optional) — range [0, 2147483647]
64	- `reference_image` (string/array, optional) — URL or base64, up to 9 images
65	- `enable_sequential` (bool, optional) — group image generation mode
66	- `thinking_mode` (bool, optional, default true) — enhanced reasoning (text-to-image only)
67	- `bbox_list` (array, optional) — bounding boxes for interactive editing
68	- `color_palette` (array, optional) — custom color theme (3-10 colors with hex+ratio)
69	- `watermark` (bool, optional, default false)
70
71	### Response
72	- `image_url` (string) — PNG, valid for 24 hours
73	- `image_count` (int)
74	- `size` (string) — actual output resolution
75	- `seed` (int)
76
77	## Quick start (Python + DashScope SDK)
78
79	```python
80	import os
81	from dashscope.aigc.image_generation import ImageGeneration
82
83	def generate_image(req: dict) -> dict:
84	messages = [
85	{
86	"role": "user",
87	"content": [{"text": req["prompt"]}],
88	}
89	]
90
91	# Add reference images if provided
92	ref_images = req.get("reference_images") or []
93	if req.get("reference_image"):
94	ref_images = [req["reference_image"]] + ref_images
95	for img in ref_images:
96	messages[0]["content"].append({"image": img})
97
98	params = {
99	"model": req.get("model", "wan2.7-image"),
100	"messages": messages,
101	"size": req.get("size", "2K"),
102	"n": req.get("n", 1),
103	"api_key": os.getenv("DASHSCOPE_API_KEY"),
104	"seed": req.get("seed"),
105	"watermark": req.get("watermark", False),
106	}
107
108	if req.get("enable_sequential"):
109	params["enable_sequential"] = True
110	if req.get("thinking_mode") is not None:
111	params["thinking_mode"] = req["thinking_mode"]
112	if req.get("bbox_list"):
113	params["bbox_list"] = req["bbox_list"]
114	if req.get("color_palette"):
115	params["color_palette"] = req["color_palette"]
116
117	response = ImageGeneration.call(**params)
118
119	content = response.output["choices"][0]["message"]["content"]
120	images = [item["image"] for item in content if isinstance(item, dict) and item.get("image")]
121
122	return {
123	"image_urls": images,
124	"image_count": response.usage.get("image_count"),
125	"size": response.usage.get("size"),
126	}
127	```
128
129	## Size reference
130
131	\| Model \| Supported sizes \| Default \|
132	\|---\|---\|---\|
133	\| wan2.7-image-pro \| 1K, 2K, 4K (text-to-image only), or [768, 4096] px \| 2K \|
134	\| wan2.7-image \| 1K, 2K, or [768, 2048] px \| 2K \|
135
136	## Error handling
137
138	\| Error \| Likely cause \| Action \|
139	\|---\|---\|---\|
140	\| 401/403 \| Missing or invalid `DASHSCOPE_API_KEY` \| Check env var or credentials file. \|
141	\| 400 `InvalidParameter` \| Unsupported size, bad n value, or missing required image \| Valida

People who liked this also viewed

6 picks

Type