# QwenLM/Qwen3-VL-Embedding

Repository: https://github.com/QwenLM/Qwen3-VL-Embedding
Canonical: https://ross.abutalabs.com/products/qwen3-vl-embedding
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2026-06-23T10:39:17+00:00

## Health v2 (maintenance only)
Score: 56/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 89, release rhythm 35, longevity 16
- inputs: {"age_days": 237, "days_push": 71, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1369, forks 121 (observed 2026-08-28T04:04:31.746240+00:00)

## What it is
Qwen3-VL-Embedding and Qwen3-VL-Reranker are state-of-the-art multimodal embedding and reranking models built on the Qwen3-VL foundation model. They process text, images, screenshots, videos, and mixed-modal inputs for information retrieval and cross-modal understanding tasks.

## Use cases
- build multimodal semantic search over images and videos
- generate embeddings for mixed text-image documents
- rerank retrieval results with vision-language relevance scoring
- implement image-text retrieval and video-text matching
- power RAG pipelines with visual document understanding
- embed screenshots for visual question answering retrieval

## When to choose
- you need state-of-the-art multimodal embeddings across text, images, and video
- you want a matched embedding plus reranker pair for a two-stage retrieval pipeline
- you are already using the Qwen model family and want consistent tooling

## When to avoid
- you only need lightweight text-only embeddings with low compute cost
- you lack GPU resources, as these are large vision-language models
- you need a plug-and-play service rather than model weights and Python code

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, search-engine, rag, nlp, computer-vision
- domain: machine-learning, large-language-models, computer-vision
- platform: python
- tags: embeddings, reranker, multimodal, vision-language-model, qwen, information-retrieval, video-understanding, retrieval-augmented-generation, search, gpu

## Member repositories
- QwenLM/Qwen3-VL-Embedding (main) score 56

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:31.746240+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T04:40:57.444264+00:00, confidence not recorded.
  - readme: https://github.com/QwenLM/Qwen3-VL-Embedding (fetched 2026-08-28T04:04:31.746240+00:00, sha 0eeb477efb1b)
- Data as of 2026-08-30T08:39:29.467469+00:00.