# OmniProvider Chat Completion (Small Tier)

> OmniProvider Chat Completion (Small Tier) is a paid API for AI agents from pay.omnidev.tech, paid per call via x402, $0.008/call, status unknown (last checked 2026-10-17).

Pay-per-call LLM chat completion on the small tier, supporting up to 8000 input and 1024 output tokens at $0.008 USDC per call on Base, with idempotent retry via client nonce.

## Facts

- Endpoint: POST https://pay.omnidev.tech/x402/chat-small?utm_source=zero.xyz
- Price: $0.008/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-17
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/omniprovider-chat-completion-small-tier-cdb09aaf
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_Zwl4eAxPzo-xwb5TJi2A0

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 omniprovider-chat-completion-small-tier-cdb09aaf -d '<json body>'
```

Example prompt: Using the small-tier model (auto-select), answer this in up to 500 tokens: 'What are the three most important factors to consider when evaluating a startup investment?'

## When to prefer this

Choose this endpoint when you need a single chat completion at low cost ($0.008 USDC) and your context fits within 8000 input tokens. Prefer this over the nano tier when you need slightly more capability or a larger context; prefer it over the large/research tiers when cost matters more than maximum intelligence. Ideal for high-volume agentic workflows where many small LLM calls are made per session and a subscription model is undesirable.

## Known failure modes

- Insufficient USDC balance or payment failure — request rejected before inference
- Input exceeds 8000 token limit — request rejected with token limit error
- max_tokens set above 1024 — schema validation error
- Neither prompt nor messages provided — missing required field error
- Invalid model enum value — schema validation error
- Network timeout before response delivered — retry with same nonce is free

## How this service works

Pay-per-call LLM chat completions and cited briefs, served via an upstream provider. MPP (Tempo USDC.e) on /v1/*, x402 v2 (Base USDC) on /x402/*. Charged only after the result is produced. /v1/chat/completions is an OpenAI-style chat completion over a multi-model catalog served via an upstream provider (JSON, or Server-Sent Events with stream:true; function tools on models whose card says capabilities.tools). Prepaid token packages are sold under /v1/packages. Restricted-access models (card field access: aup): Restricted-access models are open-weight models with their built-in refusals removed or reduced. Adults only (18+): a paid request carries the header X-Omni-Aup-Accept: 18+. Not permitted: sexual content involving minors; weapons capable of mass harm and explosives; malware, phishing and credential theft; instructions for self-harm; violence and terrorism; fraud; anything unlawful. Prompts and outputs are screened by an automated first-line filter: a matching prompt is refused before payment (400, not charged), a matching output is withheld (422, not charged). Hashes of refused requests are kept for abuse handling and repeat violations are blocked by client address. Per-call payment only (x402 or MPP), no prepaid packages, no streaming, lower daily limits than the main catalog. Policy: https://provider.omnidev.tech/docs/restricted

## Output

A single assistant message as a text completion, generated by the selected small-tier model (auto, llama-3.1-8b, or kimi-k3), up to 1024 output tokens. Payment of 0.008 USDC on Base settles after the response is produced. Retrying with the same client nonce header avoids a duplicate charge.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "model": {
   "enum": [
    "auto",
    "llama-3.1-8b",
    "kimi-k3"
   ],
   "type": "string",
   "description": "auto selects the tier model."
  },
  "prompt": {
   "type": "string",
   "description": "Shorthand for one user message; used when messages is absent."
  },
  "messages": {
   "type": "array",
   "items": {
    "type": "object",
    "required": [
     "role",
     "content"
    ],
    "properties": {
     "role": {
      "enum": [
       "system",
       "developer",
       "user",
       "assistant"
      ],
      "type": "string"
     },
     "content": {
      "type": "string",
      "description": "Message text."
     }
    },
    "additionalProperties": false
   },
   "maxItems": 64,
   "minItems": 1,
   "description": "Conversation, oldest first."
  },
  "max_tokens": {
   "type": "integer",
   "maximum": 1024,
   "minimum": 1,
   "description": "Upper bound on output tokens."
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "json",
 "example": {
  "ok": true,
  "result": {
   "id": "chatcmpl-example",
   "model": "auto",
   "usage": {
    "total_tokens": 42,
    "prompt_tokens": 18,
    "completion_tokens": 24
   },
   "object": "chat.completion",
   "choices": [
    {
     "index": 0,
     "message": {
      "role": "assistant",
      "content": "A server answers 402 with payment requirements, the client signs a payment, and the server settles it after serving."
     },
     "finish_reason": "stop"
    }
   ]
  },
  "settled": true
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/omniprovider-chat-completion-small-tier-cdb09aaf/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from pay.omnidev.tech](https://www.zero.xyz/host/pay.omnidev.tech/llms.txt)