# Agent402 Ox Alpha Chat Completions

> Agent402 Ox Alpha Chat Completions is a paid API for AI agents from agent402.tools, paid per call via x402, $0.002/call, status unknown (last checked 2026-09-15).

Serves OpenAI-compatible chat completions using the Ox Alpha reasoning model with a 1M-token context window, payable per-call via x402 USDC or free with a trial allowance.

## Facts

- Endpoint: POST https://agent402.tools/v1/ox/chat/completions
- Price: $0.002/call
- Payment: x402
- Status: unknown
- Last checked: 2026-09-15
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/agent402-ox-alpha-chat-completions-9a6668d5
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_7qYMiCPG85TQ0rtTWGVDz

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 agent402-ox-alpha-chat-completions-9a6668d5 -d '<json body>'
```

Example prompt: Use the Ox Alpha reasoning model to work through this problem step by step with high reasoning effort: given a list of 50 tasks with dependencies and deadlines, produce an optimal execution order that minimizes total latency — I need the full reasoning trace.

## When to prefer this

Choose this endpoint when you need a powerful reasoning model accessible via the standard OpenAI chat completions API format without account signup (using the trial allowance) or with per-call USDC micropayments via x402. It is ideal for agents that need a large 1M-token context window, want to avoid subscription commitments, or are operating in a pay-per-use x402 payment flow. Prefer it over standard OpenAI or Anthropic endpoints when you want zero-signup trial access or crypto-native per-call billing.

## Known failure modes

- 400 error if a model name other than stealth/ox-alpha is specified, naming which tier handles that model
- 402 Payment Required if trial allowance is exhausted and no x402 payment header is present
- Request rejected if reasoning effort 'none', 'minimal', or 'medium' is specified (not supported by this model)
- max_tokens below the floor of 1024 may be rejected or clamped
- Token limit exceeded if combined prompt and reasoning tokens exceed 1,048,576

## How this service works

OpenAI-compatible chat completions served by Ox Alpha (stealth/ox-alpha), a reasoning model with a 1,048,576-token context window. FREE TO USE while the model's own upstream is free: add ?trial=1 and no wallet, key or signup is needed (a per-client allowance, and the response says how much is left). $0.002 per call in USDC over x402 when you want it without an allowance. The model is locked to this route (sending a different model returns a 400 naming its tier).

## Output

An OpenAI-compatible chat completion response containing the assistant's message, token usage statistics (prompt, completion, reasoning tokens), and a field indicating remaining trial allowance if the trial parameter was used. The response follows the standard OpenAI ChatCompletion object format.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "model": {
   "type": "string",
   "description": "Optional - locked to stealth/ox-alpha; any other value is a 400 naming the tier that serves it."
  },
  "messages": {
   "type": "array",
   "description": "OpenAI chat messages: [{role, content}] - text and image_url content blocks supported"
  },
  "reasoning": {
   "type": "object",
   "description": "Optional - {effort: \"low\"|\"high\"|\"max\"}. Defaults to \"low\" so the budget is not spent thinking. This model always reasons; \"none\"/\"minimal\"/\"medium\" are not supported by it."
  },
  "max_tokens": {
   "type": "number",
   "description": "Output token cap (default 4096, floor 1024, tier maximum 8000). Reasoning tokens count against it, which is why the floor exists."
  },
  "cache_control": {
   "description": "Optional - prompt caching preference. Default ON ({type:\"ephemeral\"}, 5-minute TTL): repeated prefixes across your turns are served from the provider cache (same price to you). Send false to disable. ttl:\"1h\" is not offered."
  },
  "max_completion_tokens": {
   "type": "integer",
   "description": "Optional - alias of max_tokens (newer OpenAI SDKs send this)."
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "json",
 "example": {
  "id": "gen-…",
  "model": "stealth/ox-alpha",
  "usage": {
   "total_tokens": 13,
   "prompt_tokens": 12,
   "completion_tokens": 1
  },
  "object": "chat.completion",
  "choices": [
   {
    "index": 0,
    "message": {
     "role": "assistant",
     "content": "OK"
    },
    "finish_reason": "stop"
   }
  ],
  "created": 1750000000
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/agent402-ox-alpha-chat-completions-9a6668d5/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from agent402.tools](https://www.zero.xyz/host/agent402.tools/llms.txt)