aliyun-qwen-ocr

Name: aliyun-qwen-ocr
Author: cinience

59installs34forks

How to install aliyun-qwen-ocr?

cinience/alicloud-skills/aliyun-qwen-ocr

$npx -y skills add cinience/alicloud-skills --skill aliyun-qwen-ocr

Installs into the current project.

Summary

Use when OCR-specialized extraction is needed with Alibaba Cloud Model Studio Qwen OCR models (qwen-vl-ocr, qwen-vl-ocr-latest, and snapshots), including document parsing, table parsing, multilingual OCR, formula recognition, and key information extraction.

Files · 1

View on GitHub

SKILL.md

1	Category: provider
2
3	# Model Studio Qwen OCR
4
5	## Validation
6
7	```bash
8	mkdir -p output/aliyun-qwen-ocr
9	python -m py_compile skills/ai/multimodal/aliyun-qwen-ocr/scripts/prepare_ocr_request.py && echo "py_compile_ok" > output/aliyun-qwen-ocr/validate.txt
10	```
11
12	Pass criteria: command exits 0 and `output/aliyun-qwen-ocr/validate.txt` is generated.
13
14	## Output And Evidence
15
16	- Save request payloads, selected OCR task name, and normalized output expectations under `output/aliyun-qwen-ocr/`.
17	- Keep the exact model, image source, and task configuration with each saved run.
18
19	Use Qwen OCR when the task is primarily text extraction or document structure parsing rather than broad visual reasoning.
20
21	## Critical model names
22
23	Use one of these exact model strings:
24	- `qwen-vl-ocr`
25	- `qwen-vl-ocr-latest`
26	- `qwen-vl-ocr-2025-11-20`
27	- `qwen-vl-ocr-2025-08-28`
28	- `qwen-vl-ocr-2025-04-13`
29	- `qwen-vl-ocr-2024-10-28`
30
31	Selection guidance:
32	- Use `qwen-vl-ocr` for the stable channel.
33	- Use `qwen-vl-ocr-latest` only when you explicitly want the newest OCR behavior.
34	- Pin `qwen-vl-ocr-2025-11-20` when you need reproducible document parsing based on the Qwen3-VL OCR upgrade.
35
36	## Prerequisites
37
38	- Install dependencies (recommended in a venv):
39
40	```bash
41	python3 -m venv .venv
42	. .venv/bin/activate
43	python -m pip install requests
44	```
45
46	- Set `DASHSCOPE_API_KEY` in environment, or add `dashscope_api_key` to `~/.alibabacloud/credentials`.
47
48	## Normalized interface (ocr.extract)
49
50	### Request
51	- `image` (string, required): HTTPS URL, local path, or `data:` URL.
52	- `model` (string, optional): default `qwen-vl-ocr`.
53	- `prompt` (string, optional): use when you want custom extraction instructions.
54	- `task` (string, optional): built-in OCR task.
55	- `task_config` (object, optional): configuration for built-in task such as extraction fields.
56	- `enable_rotate` (bool, optional): default `false`.
57	- `min_pixels` (int, optional)
58	- `max_pixels` (int, optional)
59	- `max_tokens` (int, optional)
60	- `temperature` (float, optional): recommended to keep near default/low values.
61
62	### Response
63	- `text` (string): extracted text or structured markdown/html-style output.
64	- `model` (string)
65	- `usage` (object, optional)
66
67	## Built-in OCR tasks
68
69	Use one of these values in `task`:
70	- `text_recognition`
71	- `key_information_extraction`
72	- `document_parsing`
73	- `table_parsing`
74	- `formula_recognition`
75	- `multi_lan`
76	- `advanced_recognition`
77
78	## Quick start
79
80	Custom prompt:
81
82	```bash
83	python skills/ai/multimodal/aliyun-qwen-ocr/scripts/prepare_ocr_request.py \
84	--image "https://example.com/invoice.png" \
85	--prompt "Extract seller name, invoice date, amount, and tax number in JSON."
86	```
87
88	Built-in task:
89
90	```bash
91	python skills/ai/multimodal/aliyun-qwen-ocr/scripts/prepare_ocr_request.py \
92	--image "https://example.com/table.png" \
93	--task table_parsing \
94	--model qwen-vl-ocr-2025-11-20
95	```
96
97	## Operational guidance
98
99	- Prefer built-in OCR tasks for standard parsing jobs because they use official task prompts.
100	- For critical business fields, add downstream validation rules after OCR.
101	- `qwen-vl-ocr` and older snapshots default to `4096` max output tokens unless higher limits are approved by Alibaba Cloud; `qwen-vl-ocr-2025-11-20` follows the model maximum.
102	- Increase `max_pixels` only when small text is missed; this raises token cost.
103
104	## Output location
105
106	- Default output: `output/aliyun-qwen-ocr/request.json`
107	- Override base dir with `OUTPUT_DIR`.
108
109	## References
110
111	- `references/api_reference.md`
112	- `references/sources.md`

People who liked this also viewed

6 picks

Type

aliyun-qwen-ocr

59installs34forks

View on GitHub

How to install aliyun-qwen-ocr?

cinience/alicloud-skills/aliyun-qwen-ocr

$npx -y skills add cinience/alicloud-skills --skill aliyun-qwen-ocr

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	Category: provider
2
3	# Model Studio Qwen OCR
4
5	## Validation
6
7	```bash
8	mkdir -p output/aliyun-qwen-ocr
9	python -m py_compile skills/ai/multimodal/aliyun-qwen-ocr/scripts/prepare_ocr_request.py && echo "py_compile_ok" > output/aliyun-qwen-ocr/validate.txt
10	```
11
12	Pass criteria: command exits 0 and `output/aliyun-qwen-ocr/validate.txt` is generated.
13
14	## Output And Evidence
15
16	- Save request payloads, selected OCR task name, and normalized output expectations under `output/aliyun-qwen-ocr/`.
17	- Keep the exact model, image source, and task configuration with each saved run.
18
19	Use Qwen OCR when the task is primarily text extraction or document structure parsing rather than broad visual reasoning.
20
21	## Critical model names
22
23	Use one of these exact model strings:
24	- `qwen-vl-ocr`
25	- `qwen-vl-ocr-latest`
26	- `qwen-vl-ocr-2025-11-20`
27	- `qwen-vl-ocr-2025-08-28`
28	- `qwen-vl-ocr-2025-04-13`
29	- `qwen-vl-ocr-2024-10-28`
30
31	Selection guidance:
32	- Use `qwen-vl-ocr` for the stable channel.
33	- Use `qwen-vl-ocr-latest` only when you explicitly want the newest OCR behavior.
34	- Pin `qwen-vl-ocr-2025-11-20` when you need reproducible document parsing based on the Qwen3-VL OCR upgrade.
35
36	## Prerequisites
37
38	- Install dependencies (recommended in a venv):
39
40	```bash
41	python3 -m venv .venv
42	. .venv/bin/activate
43	python -m pip install requests
44	```
45
46	- Set `DASHSCOPE_API_KEY` in environment, or add `dashscope_api_key` to `~/.alibabacloud/credentials`.
47
48	## Normalized interface (ocr.extract)
49
50	### Request
51	- `image` (string, required): HTTPS URL, local path, or `data:` URL.
52	- `model` (string, optional): default `qwen-vl-ocr`.
53	- `prompt` (string, optional): use when you want custom extraction instructions.
54	- `task` (string, optional): built-in OCR task.
55	- `task_config` (object, optional): configuration for built-in task such as extraction fields.
56	- `enable_rotate` (bool, optional): default `false`.
57	- `min_pixels` (int, optional)
58	- `max_pixels` (int, optional)
59	- `max_tokens` (int, optional)
60	- `temperature` (float, optional): recommended to keep near default/low values.
61
62	### Response
63	- `text` (string): extracted text or structured markdown/html-style output.
64	- `model` (string)
65	- `usage` (object, optional)
66
67	## Built-in OCR tasks
68
69	Use one of these values in `task`:
70	- `text_recognition`
71	- `key_information_extraction`
72	- `document_parsing`
73	- `table_parsing`
74	- `formula_recognition`
75	- `multi_lan`
76	- `advanced_recognition`
77
78	## Quick start
79
80	Custom prompt:
81
82	```bash
83	python skills/ai/multimodal/aliyun-qwen-ocr/scripts/prepare_ocr_request.py \
84	--image "https://example.com/invoice.png" \
85	--prompt "Extract seller name, invoice date, amount, and tax number in JSON."
86	```
87
88	Built-in task:
89
90	```bash
91	python skills/ai/multimodal/aliyun-qwen-ocr/scripts/prepare_ocr_request.py \
92	--image "https://example.com/table.png" \
93	--task table_parsing \
94	--model qwen-vl-ocr-2025-11-20
95	```
96
97	## Operational guidance
98
99	- Prefer built-in OCR tasks for standard parsing jobs because they use official task prompts.
100	- For critical business fields, add downstream validation rules after OCR.
101	- `qwen-vl-ocr` and older snapshots default to `4096` max output tokens unless higher limits are approved by Alibaba Cloud; `qwen-vl-ocr-2025-11-20` follows the model maximum.
102	- Increase `max_pixels` only when small text is missed; this raises token cost.
103
104	## Output location
105
106	- Default output: `output/aliyun-qwen-ocr/request.json`
107	- Override base dir with `OUTPUT_DIR`.
108
109	## References
110
111	- `references/api_reference.md`
112	- `references/sources.md`

People who liked this also viewed

6 picks

Type