OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI APIs & Platforms

Stapi

Stapi is a self-hosted API server that generates text embeddings with a preloaded Sentence Transformers model and exposes them through an OpenAI-compatible embeddings endpoint.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Stapi — OpenViST project profile in AI APIs & Platforms.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Stapi when an application or integration expects an OpenAI-compatible embeddings endpoint and you want that endpoint to serve a Sentence Transformers model locally.

Who it is for

  • Developers integrating text embeddings into applications that use the OpenAI Embedding API format.
  • Teams connecting a locally hosted Sentence Transformers model to a Weaviate text2vec-openai integration.
  • Operators deploying an embedding endpoint with Docker or a supported local Python version.

Documented uses

  • Generate an embedding for text through the `/v1/embeddings` endpoint.
  • Point an OpenAI Python client at a local Stapi API base to request embeddings.
  • Configure Weaviate's text2vec-openai module to use a Stapi endpoint.
  • Preload a selected Sentence Transformers model for an instance.
Before you choose

Requirements & limits

  • Stapi documents both Docker and local Python installation; its Python instructions support versions 3.9, 3.10 and 3.11.
  • The default model is all-MiniLM-L6-v2; another supported model can be selected with the MODEL environment variable.
  • A running instance serves one preloaded model, and the request model field does not change it.
  • The repository identifies the project as Apache-2.0 licensed.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Start Stapi with Docker

Run the embeddings API in a Docker container with the default preloaded model.

Before you start

  • Docker

Steps

  1. Run the command below.

    Documented command
    docker run -p 8080:8080 -d ghcr.io/substratusai/stapi

  2. Open http://localhost:8080/docs to view the API documentation.

Expected outcome

A Stapi API server is started with port 8080 published on the host.

Guide limitations

  • The default preloaded model is all-MiniLM-L6-v2.
  • Each instance serves only its preloaded model; a different model requires another instance.
Documentation 1 ↗
Source-based practical guide

Request an embedding from a running Stapi server

Send text to the embeddings endpoint and retrieve an embedding using the documented cURL request.

Before you start

  • A running Stapi server reachable at localhost:8080.
  • cURL

Steps

  1. Send a JSON POST request to http://localhost:8080/v1/embeddings with Content-Type: application/json.

  2. Set the request body to include an input text string and the model field, as in the repository example.

Expected outcome

The request asks Stapi for an embedding of the supplied text.

Guide limitations

  • Stapi serves its preloaded model; the request's model parameter is ignored for model switching.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
17 October 2026
Original website title
substratusai/stapi · Official homepage title · retained in its source language
How we review sourcesSuggest a correction