# gpt-oss-120b Chat Completions (Telnyx Inference, x402)

> gpt-oss-120b Chat Completions (Telnyx Inference, x402) is a paid API for AI agents from x402.telnyx.com, paid per call via x402, $0.004928/call, status unknown (last checked 2026-11-17).

Provides OpenAI-compatible chat completions using the open-weight gpt-oss-120b reasoning model via Telnyx Inference, with configurable reasoning effort and per-call x402 micropayment pricing.

## Facts

- Endpoint: POST https://x402.telnyx.com/v1/chat/completions/_t/gpt-oss-120b?utm_source=zero.xyz
- Price: $0.004928/call
- Payment: x402
- Status: unknown
- Last checked: 2026-11-17
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/gpt-oss-120b-chat-completions-telnyx-inference-x402-bad0c8ff
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_q4jrNae7nwlkuLW8VsXcT

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 gpt-oss-120b-chat-completions-telnyx-inference-x402-bad0c8ff -d '<json body>'
```

Example prompt: Using the gpt-oss-120b reasoning model, answer the following complex multi-step logic problem with medium reasoning effort and up to 4096 tokens: 'If a train leaves Chicago at 60 mph and another leaves New York at 80 mph, and the cities are 790 miles apart, when and where do they meet?'

## When to prefer this

Choose this endpoint when you need a large 120B open-weight reasoning model with controllable thinking depth (low/medium/high), OpenAI-compatible message format, tool-call support, and transparent per-call micropayment pricing via x402 — especially for complex multi-step reasoning, agentic loops, or when you want to avoid subscription commitments and only pay for what you use.

## Known failure modes

- 503 returned when reasoning effort exhausts the entire max_tokens budget (not charged)
- Input message array exceeding 32 items or 32,000 chars per message rejected
- max_tokens outside 2048–8192 range causes validation error
- Invalid reasoning_effort value other than low/medium/high
- Payment failure via x402 protocol if insufficient USDC balance
- Model identifier mismatch — must be openai/gpt-oss-120b

## How this service works

Open-weight reasoning-model chat completions via Telnyx Inference (OpenAI gpt-oss-120b, beta). Choose reasoning_effort low (default), medium or high. Reasoning tokens count toward max_tokens (2048-8192, default 4096); if reasoning uses the whole budget you get a 503 and are not charged. OpenAI-compatible. Up to $0.004928 USDC per call, quoted to your request (input + max_tokens at list rates, no markup), or pay per-token via the upto scheme.

## Output

Returns an OpenAI-compatible chat completion response containing the assistant's reply text, optional reasoning/thinking tokens (counted against max_tokens), and any tool calls the model decides to make. If reasoning tokens consume the entire max_tokens budget, a 503 is returned and no charge is applied. Pricing is quoted upfront per call based on input length plus max_tokens.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "n": {
   "type": "integer",
   "const": 1,
   "description": "Fixed to one completion per paid call."
  },
  "model": {
   "type": "string",
   "const": "openai/gpt-oss-120b",
   "description": "Model identifier; fixed to openai/gpt-oss-120b on this route."
  },
  "tools": {
   "type": "array",
   "items": {
    "type": "object"
   },
   "maxItems": 16
  },
  "messages": {
   "type": "array",
   "items": {
    "type": "object",
    "required": [
     "role"
    ],
    "properties": {
     "role": {
      "enum": [
       "system",
       "user",
       "assistant",
       "tool"
      ],
      "type": "string"
     },
     "content": {
      "type": [
       "string",
       "null"
      ],
      "maxLength": 32000,
      "description": "Required, except null/omitted on assistant messages that carry tool_calls."
     },
     "reasoning": {
      "type": "string",
      "maxLength": 32000,
      "description": "Assistant messages only: the reasoning returned with that turn, sent back on multi-turn and tool-call loops. Counts toward the message character limits."
     },
     "reasoning_content": {
      "type": "string",
      "maxLength": 32000,
      "description": "Assistant messages only: alternative name for reasoning (as returned by GLM and gpt-oss)."
     }
    }
   },
   "maxItems": 32,
   "description": "OpenAI-style chat messages."
  },
  "max_tokens": {
   "type": "integer",
   "maximum": 8192,
   "minimum": 2048
  },
  "temperature": {
   "type": "number",
   "maximum": 2,
   "minimum": 0
  },
  "reasoning_effort": {
   "enum": [
    "low",
    "medium",
    "high"
   ],
   "type": "string",
   "default": "low",
   "description": "Thinking depth: low (default), medium or high. Every level fits the default max_tokens; high with a max_tokens near the minimum can use the whole budget (then 503, not charged)."
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/gpt-oss-120b-chat-completions-telnyx-inference-x402-bad0c8ff/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from x402.telnyx.com](https://www.zero.xyz/host/x402.telnyx.com/llms.txt)
