# AgentIntel LLM Provider Directory

> AgentIntel LLM Provider Directory is a paid API for AI agents from intel.memoryapi.org, paid per call via x402, $0.001/call, status unknown (last checked 2026-09-15).

Lists and filters LLM providers with agent-specific metrics, pricing, capability data, and tool-calling scores

## Facts

- Endpoint: GET https://intel.memoryapi.org/x402/intel/providers
- Price: $0.001/call
- Payment: x402
- Status: unknown
- Last checked: 2026-09-15
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/intel-memoryapi-org-807577b0
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_fzf3ezrDy1LSGbypp4Eah

Status and success rate cover calls made through Zero and Zero's own probes. Third-party monitors may report differently.

## How to call it through Zero

Zero handles the 402 payment challenge and records the run. With the Zero CLI installed (`npm i -g @zeroxyz/cli`):

```sh
zero fetch --capability intel-memoryapi-org-807577b0
```

Example prompt: Show me LLM providers that are best for tool calling, support at least 32k context, and cost no more than $2 per million tokens — I want to compare their agent scores and first-token latency.

## When to prefer this

Use this endpoint when an AI agent needs to programmatically select or compare LLM providers based on agent-specific metrics like tool_calling_score, agent_score, pricing, and context window — especially in agentic workflows where model selection is dynamic or cost-sensitive. Prefer over generic model comparison sites when you need machine-readable, agent-optimized scoring data.

## Known failure modes

- No providers match the filter criteria — returns count:0 with empty providers array
- Invalid filter values (e.g. non-numeric max_price) may return 400 error
- Payment failure via x402 protocol blocks access
- Stale data if data_updated timestamp is old
- Network timeout if upstream data source is unavailable

## How this service works

x402-powered LLM provider intelligence API. Get recommendations, capability data, and status for AI agent routing decisions. Pay per query in USDC on Base mainnet.

## Output

A JSON object containing a count of matching providers, data freshness timestamp, and an array of provider records each including model name, slug, company, active status, input/output pricing per million tokens, agent_score (0-100), tool_calling_score, first_token latency in ms, context window size, and capability flags (json_mode, streaming, x402_native, tool_calling).

## Example request

```json
{
 "best_for": "tool_calling",
 "max_price": 2,
 "min_context": 32000,
 "tool_calling": "true"
}
```

## Response schema (JSON Schema)

```json
{
 "example": {
  "count": 11,
  "success": true,
  "providers": [
   {
    "name": "Claude 3.5 Sonnet",
    "slug": "anthropic-claude35-sonnet",
    "pricing": {
     "input_per_million": 3
    },
    "agent_score": 91,
    "capabilities": {
     "json_mode": true,
     "x402_native": false,
     "tool_calling": true
    }
   }
  ]
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/intel-memoryapi-org-807577b0/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from intel.memoryapi.org](https://www.zero.xyz/host/intel.memoryapi.org/llms.txt)