Skip to content
runcrate
Console

AI AGENTS API

Build AI agents, one API.

Give a model your tools and let it decide when to call them. DeepSeek, Qwen, gpt-oss and Llama take the OpenAI tools format on one API.

Get an API keySee model prices

Hand this to your agent

Add a Runcrate model to this project: run curl -fsSL"https://runcrate.ai/agents.md?intent=build"and follow it.

Your agent asks for an API key. Create a key

Works inClaude CodeCodexOpenCodeClineRoo CodeKilo CodeCursor
ModelsDeepSeek, Llama, gpt-oss
FormatOpenAI tools
Parallel callsSupported

Quick start

Hand it to your agent. Or read the request it sends.

Your agent writes and runs it. The same key works for every model and every GPU.

Docs
Write a script where DeepSeek V3 on Runcrate calls our get_weather function, and run it for Tokyo. Runs curl https://api.runcrate.ai/v1/chat/completions \ -H "Authorization: Bearer rc_live_YOUR_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "model": "deepseek-ai/DeepSeek-V3", "messages": [ {"role": "user", "content": "What is the weather in Tokyo?"} ], "tools": [{ "type": "function", "function": { "name": "get_weather", "description": "Get current weather for a city", "parameters": { "type": "object", "properties": {"city": {"type": "string"}}, "required": ["city"] } } }] }'.

Available models

Models you can call today. Billed prices, as charged.

ModelMakerPriceDetail
deepseek-ai/DeepSeek-V4-ProDeepSeek$1.37 in, $2.73 out per 1M tokensReliable function calling, 1M context
meta-llama/Llama-4-Scout-17B-16E-InstructMeta$0.105 in, $0.315 out per 1M tokensMoE, efficient agent loops
See every model price

Why Runcrate

Built for production. One key, one balance.

  • OpenAI tools format

    Standard OpenAI function calling format. Define tools with JSON Schema, the model decides when and how to call them. Works with LangChain, CrewAI, and every OpenAI-compatible agent framework.

  • Parallel tool calls

    Models can request multiple tool calls in a single response. Execute them concurrently for faster agent loops and reduced round trips.

  • Multi-model agents

    Use different models for different agent tasks. DeepSeek for reasoning, DeepSeek V4 Pro for long agent runs, Llama for fast tool dispatch. One API key covers all.

  • Works with MCP clients

    An agent framework that speaks MCP can hand its tools to these models through the standard tools parameter.

FAQ

Common questions.