# Gambit GPU Shop – x402 Chat Inference

> Gambit GPU Shop – x402 Chat Inference is a paid API for AI agents from spark-2f53.tail8a7863.ts.net, paid per call via x402, $0.02/call, status unknown (last checked 2026-09-14).

Runs a paid LLM chat completion against GPU-hosted models on the Gambit platform, billed at $0.02 USDC per call via x402 on Base

## Facts

- Endpoint: POST https://spark-2f53.tail8a7863.ts.net/gambit/x402/chat
- Price: $0.02/call
- Payment: x402
- Status: unknown
- Last checked: 2026-09-14
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/gambit-gpu-shop-x402-chat-inference-3f93c94b
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_DwzOHmx6hXvERvAsrhHTQ

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 gambit-gpu-shop-x402-chat-inference-3f93c94b -d '<json body>'
```

Example prompt: Ask the Gambit GPU shop to run a chat completion using qwen3.6:35b-a3b — send it the message 'Explain quantum entanglement in two sentences' with a max of 256 tokens and pay the $0.02 USDC fee.

## When to prefer this

Choose this endpoint when you need GPU-accelerated LLM chat completions with per-call USDC micropayment billing via the x402 protocol on Base, especially when working with open-source models like Qwen hosted on a self-operated GPU shop. Prefer it over centralized API providers when you want crypto-native, pay-per-use inference without subscription accounts.

## Known failure modes

- Payment failure: x402 USDC payment on Base not authorized or insufficient balance — returns 402 Payment Required
- Invalid model name: model string does not match a hosted model — likely 4xx or model-not-found error
- Malformed messages array: missing required role or content fields — 400 Bad Request
- Token limit exceeded or inference timeout on GPU — possible 5xx or timeout response
- Network unreachability of the Tailscale host — connection refused or DNS failure

## How this service works

Gambit GPU shop. Pay with L402 Lightning on /l402/* or x402 USDC on Base on /x402/*. Handwritten letter POST /x402/letter at $25.00 USDC. Funnel https://spark-2f53.tail8a7863.ts.net/gambit. Cheap probe GET|POST /x402/models at $0.01 USDC.

## Output

Returns a JSON object containing the model name used and an assistant message object with role and content fields — the generated text reply from the LLM.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "model": {
   "type": "string"
  },
  "messages": {
   "type": "array",
   "items": {
    "type": "object",
    "required": [
     "role",
     "content"
    ],
    "properties": {
     "role": {
      "type": "string"
     },
     "content": {
      "type": "string"
     }
    }
   }
  },
  "max_tokens": {
   "type": "integer",
   "default": 512
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "json",
 "example": {
  "model": "qwen3.6:35b-a3b",
  "message": {
   "role": "assistant",
   "content": "Hi."
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/gambit-gpu-shop-x402-chat-inference-3f93c94b/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from spark-2f53.tail8a7863.ts.net](https://www.zero.xyz/host/spark-2f53.tail8a7863.ts.net/llms.txt)