# deepseek-ai/DeepSeek-VL

DeepSeek-VL: Towards Real-World Vision-Language Understanding

Repository: https://github.com/deepseek-ai/DeepSeek-VL
Canonical: https://ross.abutalabs.com/products/deepseek-vl
Homepage: https://huggingface.co/spaces/deepseek-ai/DeepSeek-VL-7B
Language: Python
License: MIT
License Family: permissive
Topics: vision-language-model, vision-language-pretraining, foundation-models
Last push: 2024-04-24T05:01:06+00:00

## Health v2 (maintenance only)
Score: 25/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 64
- inputs: {"age_days": 909, "days_push": 861, "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 4175, forks 596 (observed 2026-08-28T04:08:38.214194+00:00)

## What it is
DeepSeek-VL is an open-source vision-language foundation model for real-world multimodal understanding, released with model weights and inference code. It handles natural images, documents, web pages, formulas, and diagrams via a 7B-parameter model.

## Use cases
- answer questions about images with a vision-language model
- parse documents and web page screenshots
- recognize formulas and scientific figures
- run a multimodal LLM locally
- build image captioning or visual QA pipelines

## When to choose
- you need an open-weights multimodal model for image+text understanding
- you want to run vision-language inference on your own GPU
- you need document, diagram, or formula understanding from images

## When to avoid
- you need a maintained project with active updates
- you need text-only LLM features
- you lack GPU resources for a 7B model

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, nlp, computer-vision, llm-inference
- domain: artificial-intelligence, large-language-models, computer-vision, deep-learning
- platform: python
- tags: vision-language-model, multimodal, foundation-model, image-understanding, open-weights, natural-language-processing, gpu, linux

## Member repositories
- deepseek-ai/DeepSeek-VL (main) score 25

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:38.214194+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 2026-08-29T18:22:42.299937+00:00, confidence not recorded.
  - readme: https://github.com/deepseek-ai/DeepSeek-VL (fetched 2026-08-28T04:08:38.214194+00:00, sha a59b6941089f)
- Data as of 2026-08-30T08:39:29.467469+00:00.