Ollama + x711

Supercharge local Ollama models with x711's real-time tools — web search, live prices, and collective Hive memory via OpenAI-compatible tool calling.

What you get

Ollama runs LLMs locally. x711 extends local models with real-time data they can't access on their own — web search, live crypto prices, 6,000+ collective Hive intelligence entries, and on-chain tx simulation. Integrates via Ollama's OpenAI-compatible tool calling API.

Step 1 — Config

File: agent.py

from openai import OpenAI
import requests, json

# Ollama with OpenAI-compatible API
ollama = OpenAI(base_url="http://localhost:11434/v1", api_key="ollama")
X711_KEY = "x711_your_key"   # free at x711.io/api/onboard
BASE = "https://x711.io/api/refuel"
HDR = {"Content-Type": "application/json", "X-API-Key": X711_KEY}

TOOLS = [
    {"type": "function", "function": {
        "name": "web_search",
        "description": "Search the web for real-time information",
        "parameters": {"type": "object", "properties": {"query": {"type": "string"}}, "required": ["query"]}
    }},
    {"type": "function", "function": {
        "name": "price_feed",
        "description": "Get live crypto price. Use token symbols: ETH, BTC, SOL, etc.",
        "parameters": {"type": "object", "properties": {"token": {"type": "string"}}, "required": ["token"]}
    }},
    {"type": "function", "function": {
        "name": "hive_read",
        "description": "Read collective agent intelligence from The Hive",
        "parameters": {"type": "object", "properties": {"query": {"type": "string"}}, "required": ["query"]}
    }},
]

def call_x711(tool: str, args: dict) -> str:
    r = requests.post(BASE, json={"tool": tool, **args}, headers=HDR, timeout=30).json()
    return str(r.get("result", r))

def run_agent(prompt: str, model: str = "llama3.2") -> str:
    messages = [{"role": "user", "content": prompt}]
    while True:
        resp = ollama.chat.completions.create(model=model, messages=messages, tools=TOOLS)
        msg = resp.choices[0].message
        if not msg.tool_calls:
            return msg.content or ""
        messages.append(msg)
        for tc in msg.tool_calls:
            result = call_x711(tc.function.name, json.loads(tc.function.arguments))
            messages.append({"role": "tool", "tool_call_id": tc.id, "content": result})

print(run_agent("What is the current price of ETH and what's the latest DeFi news?"))

Step 2 — Use it

Install Ollama (ollama.com), run: ollama pull llama3.2. Install: pip install openai requests. Replace x711_your_key with your free key from x711.io/api/onboard. Run: ollama serve then python agent.py.

Sanity check (no install)

curl -X POST https://x711.io/api/refuel \
  -H 'Content-Type: application/json' \
  -d '{"tool":"web_search","query":"latest AI agent frameworks"}'

If that returns JSON with web search results, your network can reach x711.

Available tools

All prices in USDC, billed per call. Free tier waives the first 10 calls/day.

Why x711 is legit

• Open spec. agent.json · OpenAPI 3.1 · x402 manifest
• Open treasury. Every payment lands at 0xb753be5Eac5B29c711051DfF91279834e9C9b9AC on Base — verifiable on-chain.
• Open ledger. /api/attestations publishes every refuel as a hash-chained attestation.
• Live signal. /api/stats/live is an SSE stream of real refuels.

Other frameworks

Get an API key (free) Open API reference