◆ Part of Autonomous AI

One API.
Every model.

The OpenAI-compatible inference API for the Autonomous AI ecosystem. Point any OpenAI SDK at one base URL, drop in your key, and ship: streaming, tool calls, wallet-native credits, no lock-in.

Get your API key Read the quickstart
base_url  https://autonomousintelligence.io/v1
01 · Quickstart

Running in 30 seconds.

It's OpenAI-compatible. Change the base URL and key, the rest of your code stays the same. Create a key →

cURL Python Node

  
02 · Models

Three models, one endpoint.

Every model is served through a resilient multi-provider fleet with automatic failover, you get uptime, we handle routing. Common OpenAI ids (like gpt-4o) alias to the closest Jarvis model, so existing code just works.

ModelBest forContext
jarvis-70bFlagship general model (Llama 3.3 70B class). The default.32K
jarvis-8bFast & cheap for high-throughput / simple tasks.128K
jarvis-120bLargest reasoning model for deep tasks.32K
03 · API reference

Endpoints.

POST /v1/chat/completions

Create a chat completion. Set stream:true for token-by-token SSE. Supports temperature, top_p, max_tokens, stop, tools / tool_choice, and response_format.

GET /v1/models

List the available models in the standard OpenAI list shape. Requires your API key.

Example response

    
◆ Agents · DEX execution

AI Agentstrade through our DEX.

Give any AI agent the power to discover, safety-check, quote and execute Solana swaps, earning you the built-in platform fee on every trade. Connect over MCP in one line, or call the REST rail directly. It's strictly non-custodial: swap/build returns an unsigned transaction, your agent signs it locally, and swap/submit only ever receives an already-signed tx. Keys never leave your environment.

Step 1, add the MCP server. Requires Node 18+. Discovery tools (quote, safety, gems, token info) work with no key; add a key only for swaps.

Claude Code · one line
claude mcp add autonomous-intelligence \
  -e AI_AGENT_KEY=YOUR_KEY \
  -- npx -y autonomousintelligence-mcp
ChatGPT · web · any HTTP client

Add the hosted (streamable-HTTP) server:

https://autonomousintelligence.io/mcp
Claude Desktop · Cursor · Windsurf · VS Code (Cline / Continue)

Paste this into your client's MCP config, then fully restart the app:

{ "mcpServers": { "autonomous-intelligence": {
  "command": "npx",
  "args": ["-y", "autonomousintelligence-mcp"],
  "env": { "AI_AGENT_KEY": "YOUR_KEY" }
} } }

Where the config livesClaude Desktop: Settings → Developer → Edit Config (claude_desktop_config.json) · Cursor: ~/.cursor/mcp.json · Windsurf: ~/.codeium/windsurf/mcp_config.json · VS Code (Cline): the extension's MCP Servers settings. On Windows, if npx isn't found use "command": "npx.cmd". Omit the env block entirely to run discovery-only (no key).

Step 2 · Get a key (only for swaps)

Self-serve: no signup, no wallet, no email. Name your agent, click once, then set the ai_live_ key as AI_AGENT_KEY in the config above. (Discovery tools need no key.)

Prefer the API? Register with one curl
curl -s -X POST https://autonomousintelligence.io/v1/agent/register \
  -H 'content-type: application/json' \
  -d '{"agent_name":"my-agent"}'

Standalone key page (shareable, supports ?ref=): /keys. Full endpoint reference in the OpenAPI spec n/a /agent/quote, /agent/safety, /agent/gems, /agent/token, /agent/tokens are open; /agent/swap/build and /agent/swap/submit require your key.

Step 3 · Verify it works

Fully restart your client, then ask your agent "list the tradeable tokens" or "quote 0.1 SOL to USDC and rug-check the output." Or hit the rail directly:

curl -s https://autonomousintelligence.io/v1/agent/tokens
No MCP? · REST + SDKs

Call the REST rail directly (base /v1/agent n/a quote · safety · gems · token · tokens · swap/build · swap/submit), or drop in a zero-config SDK for LangChain · Vercel AI SDK · LlamaIndex · CrewAI · Eliza.

Drop-in SDKs (JS + Python) + framework adapters: /sdk · full spec: /openapi.json

04 · Authentication

API keys

Every request is authenticated with a bearer key created in your console. Keys are shown once and stored hashed, treat them like passwords, keep them server-side.

# header on every request
Authorization: Bearer jc_live_...
05 · Streaming

Server-Sent Events

With stream:true the response is text/event-stream: standard OpenAI chat.completion.chunk frames, terminated by data: [DONE]. Every OpenAI streaming client parses it out of the box.

data: {"choices":[{"delta":{"content":"Hel"}}]}
data: {"choices":[{"delta":{"content":"lo"}}]}
data: [DONE]
06 · Pricing

Pay in credits. Never expire.

Requests are metered by token against a prepaid credit balance. Start free with monthly credits, then top up by card or crypto. A request that returns no work is never charged.

STARTER
10,000
credits · $10
BUILDER · POPULAR
50,000
credits · $45
PRO
100,000
credits · $80

Business & Enterprise plans available in the console. Free monthly credits during beta.

07 · Rate limits

Fair-use limits

Free / Beta20 req/min
Premium120 req/min
Over limit429 + Retry-After
08 · Errors

OpenAI-shaped

Errors use the standard envelope, so your existing error handling works.

{"error":{"message":"...",
  "type":"invalid_request_error",
  "code":"invalid_api_key"}}

Build on Jarvis Cloud.

Create a key and make your first call in under a minute.

Get your API key →