OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Infrastructure & Serving

Open Responses Server

Open Responses Server is a self-hosted Python server that exposes Chat Completions and Responses API interfaces for AI backends, adding stateful chat and tool-call support for compatible clients.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Open Responses Server — OpenViST project profile in AI Infrastructure & Serving.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when you can configure and run a server alongside a reachable AI backend, and need an OpenAI-compatible interface for your client.

Who it is for

  • Developers connecting Codex or other OpenAI-compatible clients to self-hosted models.
  • Operators who want to route compatible API requests to backends such as Ollama, vLLM, LiteLLM or OpenAI.

Documented uses

  • Expose a reachable AI backend through Chat Completions and Responses API interfaces.
  • Connect Codex or another compatible client to an alternative model backend.
  • Handle stateful chat and tool calls, including MCP support.
Before you choose

Requirements & limits

The repository specifies an MIT license and documents installation through PyPI, from source, or with Docker.

  • You need to provide the backend's internal base URL, the server endpoint and an API key.
  • MCP use requires an MCP servers JSON configuration path.
  • Conversation history and tool-call iteration limits, streaming timeouts, server binding and logging have configurable settings.
  • Persistent state, file search, code interpreter, web search and computer-use APIs are listed as future work rather than current capabilities.

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

Install and configure the server

Set up the Python server and specify the backend and server endpoint configuration.

Before you start

  • Python package installation with pip and uv, as documented.
  • A reachable AI backend and its internal base URL.
  • Values for the server endpoint and API key.

Steps

  1. Install the package with the command below, or follow the documented uv source-install steps.

    Documented command
    pip install open-responses-server

  2. Run otc configure for interactive configuration, or set OPENAI_BASE_URL_INTERNAL, OPENAI_BASE_URL and OPENAI_API_KEY.

  3. If using MCP tools, set MCP_SERVERS_CONFIG_PATH to the MCP servers JSON file.

Expected outcome

The server has installation and backend connection settings in place; the README documents starting it separately with `otc start`.

Guide limitations

  • The guide covers installation and configuration, not a confirmed deployment or successful connection to a specific backend.
  • MCP configuration requires an MCP servers JSON file.
Documentation 1 ↗
Source-based practical guide

Check the configured server endpoint

Request the server's documented models endpoint after starting the server.

Before you start

  • The server installed and configured.
  • The server started and reachable at the configured local address.

Steps

  1. Start the installed server with otc start.

  2. Request http://localhost:8080/v1/models with curl, adjusting the address if the configured endpoint differs.

Expected outcome

A request is made to the documented models endpoint as a setup check.

Guide limitations

  • The README gives the endpoint as a setup check; it does not specify an expected response body.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
16 October 2026
Original website title
teabranch/open-responses-server · Official homepage title · retained in its source language
How we review sourcesSuggest a correction