"OpenAI Gym" matching MCP servers:

MCP server for querying OpenAI usage and cost data, including spend summaries, daily breakdowns, month-over-month comparisons, and token usage by model.

Records and scores prospective equity earnings predictions made by AI agents, allowing comparison of different agent configurations.

An MCP server for the OpenAI Ads (ChatGPT Ads) Advertiser API, providing tools for agent-led performance marketing such as finding dead ads, auditing near-duplicate copy, and auditing ad-group context hints.

Provides tools for file system operations and text generation using OpenAI-compatible models via the Model Context Protocol, supporting both stdio and HTTP transports.

Enables AI assistants to generate real images via multiple models (OpenAI, Gemini, Recraft, Seedream, Grok, Arrow) and returns usable file paths instead of base64 data.

A self-hostable, open-source, semantically-searchable Agent Skills registry delivered over MCP, with a three-tier progressive disclosure architecture.

Enables image analysis, OCR, and text-to-image generation through OpenAI-compatible APIs. Supports local paths, URLs, or base64 images with configurable models and backup endpoints.

Call OpenAI Codex from Claude Code for independent second opinions, structured code review, and delegated coding tasks through a FastMCP plugin that drives the codex CLI safely.

async/await for LLM agents: spawn and await sub-agents scheduled on your own GPUs or any OpenAI-compatible endpoint. Every tool call is a scheduled await: a resumed run goes ahead of batch work and back to the replica holding its prefix cache. Durable runs, run trees with budgets, caller fairness, a status page, OpenTelemetry.

Intelligent AI routing and integration platform for seamless provider switching

Enables image generation, editing, and composition using any OpenAI-compatible Images API within MCP clients, Dify, and n8n, delivering results as image blocks, structured fields, or resource links.

An MCP server providing vision understanding tools (image QA, OCR, video and document understanding) via OpenAI-compatible endpoints, defaulting to Zhipu GLM vision models.

A cloud vision MCP server that enables text-only LLMs like DeepSeek to analyze images via OpenAI-compatible APIs (e.g., SiliconFlow), offering a single analyze_image tool with support for URL/data URLs and automatic model fallback.

Enables querying Anytime Fitness club occupancy, hourly forecasts, nearby clubs, and personal visit history through natural language.

Enables inspecting local OpenAPI specification files by listing and retrieving spec details with filters and $ref resolution.

An MCP server that bridges Claude Code with Antigravity CLI using a Swarm Agent architecture to optimize local development workflows and minimize LLM token costs. Includes a web UI for monitoring agent workflows.

A read-only MCP server for querying AI provider administration APIs, providing normalized usage, cost, and dashboard data for OpenAI and Anthropic.

Enables ingestion and semantic search over text documents using PostgreSQL + pgvector and OpenAI-compatible embeddings, allowing any LLM agent to retrieve relevant chunks for grounded answers.

MCP server that connects LLM agents to a local LM Studio instance, enabling model management, OpenAI-compatible chat completions, text completions, and embeddings through a set of tools.

Multi-provider image generation MCP server that enables image generation from Claude Desktop, Claude Code, or any MCP client using OpenAI, Google Gemini, Stable Diffusion, or a placeholder provider.

A local web search MCP server supporting parallel, API, and model-based search modes, integrating Zhipu Web Search API and online models for real-time information retrieval.

Connects Claude-compatible clients to the Elephant data graph, enabling discovery of data groups, classes, and property schemas. Supports semantic code search for verified script examples via OpenAI or AWS Bedrock embeddings.

Enables Claude Code to delegate coding tasks—investigation, review, long-running background jobs, and optional file edits—to a locally installed OpenAI Codex CLI, with the model and reasoning effort chosen per task and read-only execution by default. Follow-up requests reuse Codex's existing context, and write-enabled delegations can be confined to a managed git worktree so the user's own checkout stays untouched.

A deep-research MCP server exposing a single task-augmented tool that runs GPT-Researcher with Tavily search, Anthropic Claude, and OpenAI embeddings, streaming progress via MCP tasks and Upjack entity stream.

Lets MCP-aware agents consult rival LLMs mid-conversation, sending questions to a specific provider or fanning out to OpenAI, Gemini, Anthropic, and xAI/Grok in parallel for second opinions.

Enables LLMs to dynamically search, scrape, and query official documentation of libraries like uv, OpenAI, LangChain, and LlamaIndex via Google Serper and Groq.

MCP server exposing pure mathematical operations (sum, subtraction, multiplication, division) as tools, with a clean layered architecture and optional OpenAI Agents SDK integration.

Discovers, inspects, and executes requests against OpenAPI or Swagger APIs directly from documentation URLs or specification files. It enables users to trace parameter usage across an API and perform authenticated HTTP requests through a structured toolset.

MCP server for audio transcription using local faster-whisper or OpenAI Whisper API, enabling multilingual transcription with optional GPT post-processing.

Proxies LLM completion requests to OpenAI-compatible providers via MCP tools.