# ZeroReader GPT-OSS 20B Chat Completion

> ZeroReader GPT-OSS 20B Chat Completion is a paid API for AI agents from api.zeroreader.com, paid per call via x402, $0.002/call, status unknown (last checked 2026-10-02).

Runs chat completions using OpenAI's open-source 20B language model via a pay-per-call x402 API

## Facts

- Endpoint: POST https://api.zeroreader.com/v1/ai/gpt-oss-20b?utm_source=zero.xyz
- Price: $0.002/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-02
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/zeroreader-gpt-oss-20b-chat-completion-7410569e
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_hqHhAVqjGvTvfKLT294PR

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 zeroreader-gpt-oss-20b-chat-completion-7410569e -d '<json body>'
```

Example prompt: Use the GPT-OSS 20B model to answer this question — set temperature to 0.7 and limit the response to 512 tokens: 'What are the key differences between transformer and state-space models?'

## When to prefer this

Choose this endpoint when you need a mid-size capable open-source model (20B parameters) that balances quality and cost at $0.002/call, and you want OpenAI's open-source architecture rather than Meta's Llama or Mistral. Prefer over GPT-OSS 120B when latency and cost matter more than peak capability, and over smaller models like Llama 3.2 3B when task complexity demands more capacity.

## Known failure modes

- Payment not provided or insufficient USDC balance — 402 Payment Required
- messages array missing or empty — 400 Bad Request
- temperature out of range (>2 or <0) — 422 Validation Error
- max_tokens exceeds 4096 — 422 Validation Error
- Model overload or upstream inference failure — 503 Service Unavailable
- Malformed message role (not system/user/assistant) — 400 Bad Request

## How this service works

GPT-OSS 20B (OpenAI) — OpenAI's open-source 20B model.

## Output

Returns a chat.completion object containing the assistant's generated reply, a finish_reason (e.g. 'stop'), a unique completion ID, and message metadata — structured identically to the OpenAI chat completion response format.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "stream": {
   "type": "boolean",
   "default": false
  },
  "messages": {
   "type": "array",
   "items": {
    "type": "object",
    "required": [
     "role",
     "content"
    ],
    "properties": {
     "role": {
      "enum": [
       "system",
       "user",
       "assistant"
      ],
      "type": "string"
     },
     "content": {
      "type": "string"
     }
    }
   }
  },
  "max_tokens": {
   "type": "integer",
   "default": 1024,
   "maximum": 4096
  },
  "temperature": {
   "type": "number",
   "default": 0.7,
   "maximum": 2,
   "minimum": 0
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "id": "chatcmpl-example",
 "object": "chat.completion",
 "choices": [
  {
   "index": 0,
   "message": {
    "role": "assistant",
    "content": "I'm doing well!"
   },
   "finish_reason": "stop"
  }
 ]
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/zeroreader-gpt-oss-20b-chat-completion-7410569e/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from api.zeroreader.com](https://www.zero.xyz/host/api.zeroreader.com/llms.txt)