# Taifoon Chat Completions – hy4-preview Model

> Taifoon Chat Completions – hy4-preview Model is a paid API for AI agents from coord.taifoon.dev, paid per call via x402, $0.0012/call, status unknown (last checked 2026-11-20).

OpenAI-compatible pay-per-call chat completion endpoint serving 71 LLMs via the Taifoon coordination layer, defaulting to NVIDIA-Nemotron-3-Nano-4B-BF16

## Facts

- Endpoint: POST https://coord.taifoon.dev/v1/chat/completions/hy4-preview?utm_source=zero.xyz
- Price: $0.0012/call
- Payment: x402
- Status: unknown
- Last checked: 2026-11-20
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/taifoon-chat-completions-hy4-preview-model-871cd6f9
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_Je7oK-ksu6srjWbVPyNpQ

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 taifoon-chat-completions-hy4-preview-model-871cd6f9 -d '<json body>'
```

Example prompt: Ask the Taifoon hy4-preview endpoint to answer this for me using the phi-4 model, up to 200 tokens: 'Explain the difference between supervised and unsupervised learning in two sentences.'

## When to prefer this

Choose this endpoint when you need cheap, pay-per-call LLM inference across a wide selection of open-weight models (71 options) without a subscription, and you want OpenAI-compatible request/response format. Especially useful for agents that need to route through different models dynamically or want micro-payment billing via x402. Prefer over OpenAI directly when cost per call matters and open-source models are acceptable.

## Known failure modes

- Missing or invalid x402 payment header returns 402 Payment Required
- messages array empty or missing causes 422 validation error
- max_tokens exceeds 700 causes schema rejection
- Invalid model name falls back to default or returns 400
- Rate limiting or quota exhaustion from the coordination layer returns 429
- Network timeout on the underlying inference provider returns 504

## How this service works

OpenAI-compatible chat completion on tencent/Hy4-preview, served by DeepInfra (serverless inference) and sold by Taifoon at the provider's price plus a 20 % reseller cut: 1.0008 / 3.0012 USD per 1M tokens in/out. This request is priced at its prompt and max_tokens. The answer is checked before you are charged.

## Output

A JSON chat completion object in OpenAI format containing: an ID, the model used, the assistant's reply message, finish reason (e.g. 'stop'), prompt/completion/total token counts, and a created timestamp. The choices array contains the generated text content.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "model": {
   "type": "string",
   "examples": [
    "l3-8b-lunaris-v1",
    "llama-3.1-8b-instruct",
    "mistral-small-24b-instruct-2501",
    "gemma-3-4b-it",
    "gpt-oss-20b",
    "phi-4",
    "qwen2.5-7b-instruct",
    "gemma-3-12b-it",
    "qwen3.5-9b",
    "gemma-3-27b-it",
    "ling-3.0-flash",
    "ling-3.0-flash-vl"
   ],
   "description": "optional; one of the 76 models this wall delivers (GET lists them under terms.models); without it nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16 answers"
  },
  "messages": {
   "type": "array",
   "items": {
    "type": "object",
    "required": [
     "role",
     "content"
    ],
    "properties": {
     "role": {
      "enum": [
       "system",
       "user",
       "assistant"
      ],
      "type": "string"
     },
     "content": {
      "type": "string"
     }
    }
   },
   "maxItems": 32,
   "minItems": 1
  },
  "max_tokens": {
   "type": "integer",
   "maximum": 700,
   "minimum": 1
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "json",
 "example": {
  "id": "chatcmpl-gpu_…",
  "model": "nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16",
  "usage": {
   "total_tokens": 28,
   "prompt_tokens": 14,
   "completion_tokens": 14
  },
  "object": "chat.completion",
  "choices": [
   {
    "index": 0,
    "message": {
     "role": "assistant",
     "content": "17 × 24 = 408 (17 × 20 = 340; 17 × 4 = 68; 340 + 68 = 408)."
    },
    "finish_reason": "stop"
   }
  ],
  "created": 1791763200
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/taifoon-chat-completions-hy4-preview-model-871cd6f9/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from coord.taifoon.dev](https://www.zero.xyz/host/coord.taifoon.dev/llms.txt)
