# AI Power User Chat Completions (Self-Hosted LLM via x402)

> AI Power User Chat Completions (Self-Hosted LLM via x402) is a paid API for AI agents from 2x-3090.tailf80b7a.ts.net, paid per call via x402, $0.01/call, status down (last checked 2026-09-15).

Sends a chat completion request to a self-hosted LLM endpoint, returning an assistant message response in OpenAI-compatible format for $0.01 USDC per call.

## Facts

- Endpoint: POST https://2x-3090.tailf80b7a.ts.net/x402/v1/chat/completions
- Price: $0.01/call
- Payment: x402
- Status: down
- Last checked: 2026-09-15
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/ai-power-user-chat-completions-self-hosted-llm-via-x402-adc7fc07
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_6udwAEgnkfdp4LBX2zY1C

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 ai-power-user-chat-completions-self-hosted-llm-via-x402-adc7fc07 -d '<json body>'
```

Example prompt: Send this message to the self-hosted chat completions endpoint and get a response: 'Explain the difference between a transformer and an RNN in two sentences.'

## When to prefer this

Choose this endpoint when you need low-cost ($0.01/call), OpenAI-compatible chat completions via micropayment (x402/USDC), especially in autonomous agent workflows where cost control matters and the OpenAI API format is already supported. Ideal for developers building on x402 payment rails or experimenting with self-hosted LLM inference accessed over the internet.

## Known failure modes

- Payment failure — insufficient USDC balance or x402 payment rejected
- Model unavailability — self-hosted node offline or Tailscale tunnel down
- Malformed request — missing or invalid messages array in request body
- Rate limiting or resource exhaustion on the local hardware
- Timeout — local model inference takes too long on available hardware

## How this service works

Master AI on your own hardware. 3 guides covering local AI models, Claude Code, and cost-saving setups. 150+ pages, PDF + DOCX instant download.

## Output

Returns a JSON object with a choices array containing the assistant's generated message, including role and content fields, in OpenAI chat completions format.

## Response schema (JSON Schema)

```json
{
 "type": "json",
 "example": {
  "choices": [
   {
    "message": {
     "role": "assistant",
     "content": "..."
    }
   }
  ]
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/ai-power-user-chat-completions-self-hosted-llm-via-x402-adc7fc07/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from 2x-3090.tailf80b7a.ts.net](https://www.zero.xyz/host/2x-3090.tailf80b7a.ts.net/llms.txt)