open-responses-server

teabranch
68
Wraps any OpenAI API interface as Responses with MCPs support so it supports Codex. Adding any missing stateful features. Ollama and Vllm compliant.
#ai #codex #generative-ai #mcp #mcp-client #openai #openai-api #openai-codex #openai-codex-cli #openai-codex-integration #responses-api

Overview

open-responses-server Introduction

open-responses-server is a plug-and-play server that interfaces with OpenAI’s Responses API, allowing compatibility with various AI backends like Ollama and vLLM. It supports Codex and adds stateful features for enhanced functionality.

How to Use

To use open-responses-server, install it via PyPI with 'pip install open-responses-server' or from source. Run the server using the CLI tool with 'otc start' or directly from the source. Docker deployment is also supported for containerized environments.

Key Features

Key features include compatibility with multiple AI backends, support for OpenAI’s Coding Assistant (Codex), stateful chat handling, tool call support, and enterprise-ready scalability and security.

Where to Use

open-responses-server can be used in various fields including software development, research, and any application requiring AI interaction via OpenAI’s API, especially in environments that utilize self-hosted LLMs.

Use Cases

Use cases include running OpenAI’s Coding Assistant against custom models, experimenting with self-hosted LLMs while maintaining OpenAI API compatibility, and developing innovative AI applications in research and enterprise settings.

Content