# jina-ai/clip-as-service

🏄 Scalable embedding, reasoning, ranking for images and sentences with CLIP

Repository: https://github.com/jina-ai/clip-as-service
Canonical: https://ross.abutalabs.com/products/clip-as-service
Homepage: https://clip-as-service.jina.ai
Language: Python
License: NOASSERTION
License Family: other
Topics: bert, sentence-encoding, deep-learning, clip-model, clip-as-service, bert-as-service, cross-modal-retrieval, multi-modality, neural-search, openai, pytorch, onnx, cross-modality, image2vec, sentence2vec
Last push: 2024-01-23T10:33:43+00:00

## Health v2 (maintenance only)
Score: 23/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 8, longevity 100
- inputs: {"age_days": 2851, "days_push": 953, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 12836, forks 2067 (observed 2026-08-28T04:10:59.888907+00:00)

## What it is
CLIP-as-service is a low-latency, high-scalability server for embedding images and text into fixed-length vectors using OpenAI's CLIP model. It exposes gRPC, HTTP, and WebSocket APIs and integrates with the Jina/DocArray neural search ecosystem.

## Use cases
- embed images and sentences into vectors with CLIP
- serve CLIP model as a scalable microservice
- build cross-modal image-text search
- rank images against text queries
- run visual reasoning with CLIP zero-shot classification
- host CLIP inference on GPU server with gRPC client

## When to choose
- you need a production-grade, low-latency CLIP embedding service with load balancing
- you want gRPC/HTTP/WebSocket clients decoupled from a GPU server
- you are building neural search with Jina or DocArray

## When to avoid
- you only need one-off local CLIP embeddings without a service layer
- you need models beyond CLIP variants
- you require actively developed features, as the project is in maintenance mode

## Facets
- artifact type: service
- maturity: maintenance
- function: machine-learning, search-engine, http-server, rpc
- domain: machine-learning, deep-learning, computer-vision, artificial-intelligence
- platform: python, self-hosted, cross-platform
- tags: clip, embeddings, multimodal, cross-modal-retrieval, image-embedding, sentence-embedding, neural-search, onnx, tensorrt, grpc, embedding, search, natural-language-processing, docker, gpu

## Member repositories
- jina-ai/clip-as-service (main) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:59.888907+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 2026-08-29T17:13:45.601592+00:00, confidence not recorded.
  - readme: https://github.com/jina-ai/clip-as-service (fetched 2026-08-28T04:10:59.888907+00:00, sha 85801c200e46)
  - homepage: https://clip-as-service.jina.ai (fetched 2026-08-29T08:09:49.441655+00:00, sha 614ca46fb4d8)
  - site_page: https://clip-as-service.jina.ai/user-guides/faq (fetched 2026-08-29T08:09:49.444575+00:00, sha ace5faa12eee)
  - registry_pypi: https://pypi.org/pypi/clip-as-service/json (fetched 2026-08-29T08:09:49.446514+00:00, sha 2335eaf4aaba)
- Data as of 2026-08-30T08:39:29.467469+00:00.