Bring Machine Learning
to OpenAI Agents SDK
Create your Vinkius account to connect Replicate to OpenAI Agents SDK and start using all 12 AI tools in minutes. Fully managed, enterprise secure, and ready to use without writing a single line of code. No hosting, no server setup — just connect and start using.
GDPR Free for SubscribersCompatible with every major AI agent and IDE
Gemini
What is the Replicate MCP Server?
Connect your conversational assistant directly to the Replicate ecosystem. This integration grants your AI the ability to interact programmatically with a vast library of open-source machine learning models without running them on your local hardware. From orchestrating complex image generations to spinning up specialized language models, you can command AI workflows directly from your chat.
What you can do
- Execute Predictions — Command the assistant to execute specific model versions on your behalf (
create_prediction) by supplying a payload of variables. Monitor long-running processes by retrieving outputs and execution status reliably (get_prediction) or cancel them at will (cancel_prediction). - Discover Models — Instruct the AI to intelligently scan the Replicate platform for models matching a specific use case using
search_models. You can also explore trending and categorized models by leveraging thelist_collectionsaction. - Analyze Model Metadata — Whenever you discover a new model, query its precise owner and name (
get_model) to extract the exact schema and parameter requirements necessary for a successful execution. You can also view a log of your own executed tasks (list_predictions).
How it works
- Install the Replicate platform extension module in your MCP.
- Provide your personal
Replicate API Token, extracted from your Replicate Account Settings panel. Deposit it securely into the configuration variables below. - Prompt your assistant organically: "Search for a high-quality video generation model, evaluate its parameter schema, and start generating a clip using the prompt 'a cat walking on Mars'."
Who is this for?
- AI Developers & Researchers — Explore and test novel open-source algorithms by generating quick predictions without modifying Python notebook code.
- Content Creators — Execute specialized image, audio, or video generation tasks by directly delegating workflows to your conversational assistant.
- Builders — Mix and match the output of various specialized models intelligently using natural language instructions.
Built-in capabilities (12)
Cancels a prediction that is currently running
g., image generation, LLMs). Provide the model version ID and inputs as a JSON object. Starts a new model prediction on Replicate
Retrieves the authenticated Replicate account details
Provide the collection slug (e.g., "text-to-image"). Retrieves a specific collection of models by its slug
Retrieves details for a specific model
). Retrieves the status and output of a prediction
g., "Image-to-Text", "Audio Generation"). Lists curated collections of models
Lists your active model deployments on Replicate
Lists available GPU hardware options for running models
Lists public models available on Replicate
Lists recent predictions made by the user
Searches for public models on Replicate
Why OpenAI Agents SDK?
The OpenAI Agents SDK auto-discovers all 12 tools from Replicate through native MCP integration. Build agents with built-in guardrails, tracing, and handoff patterns. chain multiple agents where one queries Replicate, another analyzes results, and a third generates reports, all orchestrated through Vinkius.
- —
Native MCP integration via
MCPServerSse, pass the URL and the SDK auto-discovers all tools with full type safety - —
Built-in guardrails, tracing, and handoff patterns let you build production-grade agents without reinventing safety infrastructure
- —
Lightweight and composable: chain multiple agents and MCP servers in a single pipeline with minimal boilerplate
- —
First-party OpenAI support ensures optimal compatibility with GPT models for tool calling and structured output
Replicate in OpenAI Agents SDK
Why run Replicate with Vinkius?
The Replicate connection runs on our fully managed, secure cloud infrastructure. We handle the hosting, maintenance, and security so you don't have to deal with servers or code. All 12 tools are ready to work instantly without any complex setup.
You stay in complete control of your data. Your AI only accesses the information you approve, keeping your sensitive passwords and private details completely safe. Plus, with automatic optimizations, your AI works faster and more efficiently.
* Every connection is hosted and maintained by Vinkius. We handle the security, updates, and infrastructure so you don't have to write code or manage servers. See our infrastructure
Over 4,000 integrations ready for AI agents
Explore a vast library of pre-built integrations, optimized and ready to deploy.
Connect securely in under 30 seconds
Generate tokens to authenticate and link external services in a single step.
Complete visibility into every agent action
Audit live requests, latency, success rates, and active security compliance policies.
Optimize spending and track token ROI
Analyze real-time token consumption and cost metrics detailed by connection.
Explore our live AI Agents Analytics dashboard to see it all working
This dashboard is included when you connect Replicate using Vinkius. You will never be left in the dark about what your AI agents are doing with your tools.
Replicate and 4,000+ other AI tools. No hosting, no code, ready to use.
Professionals who connect Replicate to OpenAI Agents SDK through Vinkius don't need to write code, manage servers, or worry about security. Everything is pre-configured, secure, and runs automatically in the background.
Raw MCP | Vinkius | |
|---|---|---|
| Ready-to-use MCPs | Find and configure each manually | 4,000+ MCPs ready to use |
| Connection Setup | Manual coding & server setup | 1-click instant connection |
| Server Hosting | You host it yourself (needs 24/7 uptime) | 100% hosted & managed by Vinkius |
| Security & Privacy | Stored in plaintext config files | Bank-grade encrypted vault |
| Activity Visibility | Blind execution (no logs or tracking) | Live dashboard with real-time logs |
| Cost Control | Runaway AI token spend risk | Automatic budget limits |
| Revoking Access | Must delete files or code to stop | 1-click disconnect button |
How Vinkius secures
Replicate for OpenAI Agents SDK
Every request between OpenAI Agents SDK and Replicate is protected by our secure gateway. We automatically keep your sensitive data private, prevent unauthorized access, and let you disconnect instantly at any time.
Frequently asked questions
Can the agent pass a JSON payload directly into a Replicate model?
Yes. You can utilize the create_prediction action and attach the payload parameter filled out with any required input schema (e.g., specific prompt, num_inference_steps). Since models change inputs constantly, you should always ask your assistant to fetch the schema details first via get_model to verify keys.
Does the prediction command return results instantly?
No, Replicate's API operates asynchronously. The initial command gives your assistant an ID. You must then ask your AI companion to query the get_prediction tool periodically using that generated ID until it displays the completed status along with the generated web URLs or generated strings.
Can the AI browse trending or curated model collections?
Yes. Use the list_collections tool to browse curated groups of models organized by category — such as image generation, text-to-speech, or video. Each collection includes a slug and description so you can quickly identify the right set of models for your use case.
How does the OpenAI Agents SDK connect to MCP?
Use MCPServerSse(url=...) to create a server connection. The SDK auto-discovers all tools and makes them available to your agent with full type information.
Can I use multiple MCP servers in one agent?
Yes. Pass a list of MCPServerSse instances to the agent constructor. The agent can use tools from all connected servers within a single run.
Does the SDK support streaming responses?
Yes. The SDK supports SSE and Streamable HTTP transports, both of which work natively with Vinkius.
MCPServerStreamableHttp not found
Ensure you have the latest version: pip install --upgrade openai-agents
Agent not calling tools
Make sure your prompt explicitly references the task the tools can help with.
Explore More MCP Servers
View all →
Ragas
7 toolsEquip your AI with Ragas to create datasets, run RAG evaluations, and track experiment metrics directly from your workflow.
Vapi
10 toolsCommand Voice AI assistants directly from your chat. Make outbound phone calls, update personas, and retrieve full transcripts via Vapi.
Twilio SMS Sender
1 toolsThis MCP does exactly one thing: it sends raw SMS messages using Twilio. That's its only function, and nothing else. Incredible for giving your AI agents out-of-band alerting capabilities.
DeepL
9 toolsTranslate and audit text — manage multilingual communication via AI.