# SYNTHORA Inference Router

> SYNTHORA Inference Router is a paid API for AI agents from inference.hergertsynthora.com, paid per call via x402, $0.02/call, status unknown (last checked 2026-09-13).

Pay-per-prompt LLM inference using a sovereign constitution-tuned SYNTHORA-Qwen model with Ed25519-signed outputs, no API key required.

## Facts

- Endpoint: POST https://inference.hergertsynthora.com/service
- Price: $0.02/call
- Payment: x402
- Status: unknown
- Last checked: 2026-09-13
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/synthora-inference-router-677f3e12
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_A4R5p2hO3AZ5timkbC2hf

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 synthora-inference-router-677f3e12 -d '<json body>'
```

Example prompt: Ask the SYNTHORA-Qwen model to write a short professional bio for a software engineer named Alex Chen who specializes in distributed systems, using the system prompt 'You are a concise professional copywriter' and limit the response to 150 tokens.

## When to prefer this

Choose this endpoint when you need LLM inference without managing API keys or OAuth flows, want cryptographically verifiable (Ed25519-signed) AI outputs, are comfortable with x402 pay-per-prompt micropayments in USDC, or specifically need the SYNTHORA constitution-tuned Qwen model's behavior rather than a generic hosted LLM.

## Known failure modes

- Payment not received or insufficient USDC balance — 402 Payment Required
- Prompt or system field missing or malformed JSON — 400 Bad Request
- Model inference timeout for very long prompts — 504 Gateway Timeout
- max_tokens value out of acceptable range — 400 validation error
- Service unavailable or model overloaded — 503 Service Unavailable

## How this service works

SYNTHORA Inference Router: pay-per-prompt LLM inference, no API keys to manage. Sovereign constitution-tuned SYNTHORA-Qwen model, local, Ed25519-signed output. POST {"prompt":"...","system":"...","max_tokens":200}.

## Output

Returns an LLM-generated text completion from the SYNTHORA-Qwen model, cryptographically signed with Ed25519, verifiable for authenticity. The response reflects the prompt and optional system prompt, truncated at the specified max_tokens limit.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "prompt": {
   "type": "string",
   "description": "The prompt to run"
  },
  "system": {
   "type": "string",
   "description": "Optional system prompt"
  },
  "max_tokens": {
   "type": "integer",
   "description": "Max output tokens (default 400)"
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "json",
 "schema": {
  "type": "object",
  "required": [
   "ok",
   "niche"
  ],
  "properties": {
   "ok": {
    "type": "boolean"
   },
   "error": {
    "type": "string"
   },
   "niche": {
    "type": "string"
   },
   "result": {
    "type": "object",
    "properties": {
     "note": {
      "type": "string"
     },
     "text": {
      "type": "string"
     },
     "tier": {
      "type": "string"
     },
     "brain": {
      "type": "string"
     },
     "model": {
      "type": "string"
     },
     "latency_ms": {
      "type": "number"
     },
     "price_usdc": {
      "type": "number"
     }
    }
   },
   "receipt": {
    "type": "object"
   }
  }
 },
 "example": {
  "ok": true,
  "niche": "inference",
  "result": {
   "text": "x402 is an HTTP-native micropayment protocol…",
   "signed": "ed25519"
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/synthora-inference-router-677f3e12/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from inference.hergertsynthora.com](https://www.zero.xyz/host/inference.hergertsynthora.com/llms.txt)