aliyun-qwen-multimodal-embedding

Name: aliyun-qwen-multimodal-embedding
Author: cinience

50installs34forks

How to install aliyun-qwen-multimodal-embedding?

cinience/alicloud-skills/aliyun-qwen-multimodal-embedding

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

Installs into the current project.

Summary

Use when multimodal embeddings are needed from Alibaba Cloud Model Studio models such as qwen3-vl-embedding for image, video, and text retrieval, cross-modal search, clustering, or offline vectorization pipelines.

Files · 1

View on GitHub

SKILL.md

1	Category: provider
2
3	# Model Studio Multimodal Embedding
4
5	## Validation
6
7	```bash
8	mkdir -p output/aliyun-qwen-multimodal-embedding
9	python -m py_compile skills/ai/search/aliyun-qwen-multimodal-embedding/scripts/prepare_multimodal_embedding_request.py && echo "py_compile_ok" > output/aliyun-qwen-multimodal-embedding/validate.txt
10	```
11
12	Pass criteria: command exits 0 and `output/aliyun-qwen-multimodal-embedding/validate.txt` is generated.
13
14	## Output And Evidence
15
16	- Save normalized request payloads, selected dimensions, and sample input references under `output/aliyun-qwen-multimodal-embedding/`.
17	- Record the exact model, modality mix, and output vector dimension for reproducibility.
18
19	Use this skill when the task needs text, image, or video embeddings from Model Studio for retrieval or similarity workflows.
20
21	## Critical model names
22
23	Use one of these exact model strings as needed:
24	- `qwen3-vl-embedding`
25	- `qwen2.5-vl-embedding`
26	- `tongyi-embedding-vision-plus-2026-03-06`
27
28	Selection guidance:
29	- Prefer `qwen3-vl-embedding` for the newest multimodal embedding path.
30	- Use `qwen2.5-vl-embedding` when you need compatibility with an older deployed pipeline.
31
32	## Prerequisites
33
34	- Set `DASHSCOPE_API_KEY` in your environment, or add `dashscope_api_key` to `~/.alibabacloud/credentials`.
35	- Pair this skill with a vector store such as DashVector, OpenSearch, or Milvus when building retrieval systems.
36
37	## Normalized interface (embedding.multimodal)
38
39	### Request
40	- `model` (string, optional): default `qwen3-vl-embedding`
41	- `texts` (array<string>, optional)
42	- `images` (array<string>, optional): public URLs or local paths uploaded by your client layer
43	- `videos` (array<string>, optional): public URLs where supported
44	- `dimension` (int, optional): e.g. `2560`, `2048`, `1536`, `1024`, `768`, `512`, `256` for `qwen3-vl-embedding`
45
46	### Response
47	- `embeddings` (array<object>)
48	- `dimension` (int)
49	- `usage` (object, optional)
50
51	## Quick start
52
53	```bash
54	python skills/ai/search/aliyun-qwen-multimodal-embedding/scripts/prepare_multimodal_embedding_request.py \
55	--text "A cat sitting on a red chair" \
56	--image "https://example.com/cat.jpg" \
57	--dimension 1024
58	```
59
60	## Operational guidance
61
62	- Keep `input.contents` as an array; malformed shapes are a common 400 cause.
63	- Pin the output dimension to match your index schema before writing vectors.
64	- Use the same model and dimension across one vector index to avoid mixed-vector incompatibility.
65	- For large image or video batches, stage files in object storage and reference stable URLs.
66
67	## Output location
68
69	- Default output: `output/aliyun-qwen-multimodal-embedding/request.json`
70	- Override base dir with `OUTPUT_DIR`.
71
72	## References
73
74	- `references/sources.md`

People who liked this also viewed

6 picks

Type

aliyun-qwen-multimodal-embedding

50installs34forks

View on GitHub

How to install aliyun-qwen-multimodal-embedding?

cinience/alicloud-skills/aliyun-qwen-multimodal-embedding

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

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	Category: provider
2
3	# Model Studio Multimodal Embedding
4
5	## Validation
6
7	```bash
8	mkdir -p output/aliyun-qwen-multimodal-embedding
9	python -m py_compile skills/ai/search/aliyun-qwen-multimodal-embedding/scripts/prepare_multimodal_embedding_request.py && echo "py_compile_ok" > output/aliyun-qwen-multimodal-embedding/validate.txt
10	```
11
12	Pass criteria: command exits 0 and `output/aliyun-qwen-multimodal-embedding/validate.txt` is generated.
13
14	## Output And Evidence
15
16	- Save normalized request payloads, selected dimensions, and sample input references under `output/aliyun-qwen-multimodal-embedding/`.
17	- Record the exact model, modality mix, and output vector dimension for reproducibility.
18
19	Use this skill when the task needs text, image, or video embeddings from Model Studio for retrieval or similarity workflows.
20
21	## Critical model names
22
23	Use one of these exact model strings as needed:
24	- `qwen3-vl-embedding`
25	- `qwen2.5-vl-embedding`
26	- `tongyi-embedding-vision-plus-2026-03-06`
27
28	Selection guidance:
29	- Prefer `qwen3-vl-embedding` for the newest multimodal embedding path.
30	- Use `qwen2.5-vl-embedding` when you need compatibility with an older deployed pipeline.
31
32	## Prerequisites
33
34	- Set `DASHSCOPE_API_KEY` in your environment, or add `dashscope_api_key` to `~/.alibabacloud/credentials`.
35	- Pair this skill with a vector store such as DashVector, OpenSearch, or Milvus when building retrieval systems.
36
37	## Normalized interface (embedding.multimodal)
38
39	### Request
40	- `model` (string, optional): default `qwen3-vl-embedding`
41	- `texts` (array<string>, optional)
42	- `images` (array<string>, optional): public URLs or local paths uploaded by your client layer
43	- `videos` (array<string>, optional): public URLs where supported
44	- `dimension` (int, optional): e.g. `2560`, `2048`, `1536`, `1024`, `768`, `512`, `256` for `qwen3-vl-embedding`
45
46	### Response
47	- `embeddings` (array<object>)
48	- `dimension` (int)
49	- `usage` (object, optional)
50
51	## Quick start
52
53	```bash
54	python skills/ai/search/aliyun-qwen-multimodal-embedding/scripts/prepare_multimodal_embedding_request.py \
55	--text "A cat sitting on a red chair" \
56	--image "https://example.com/cat.jpg" \
57	--dimension 1024
58	```
59
60	## Operational guidance
61
62	- Keep `input.contents` as an array; malformed shapes are a common 400 cause.
63	- Pin the output dimension to match your index schema before writing vectors.
64	- Use the same model and dimension across one vector index to avoid mixed-vector incompatibility.
65	- For large image or video batches, stage files in object storage and reference stable URLs.
66
67	## Output location
68
69	- Default output: `output/aliyun-qwen-multimodal-embedding/request.json`
70	- Override base dir with `OUTPUT_DIR`.
71
72	## References
73
74	- `references/sources.md`

People who liked this also viewed

6 picks

Type