# AgentRXI Second Opinion

> AgentRXI Second Opinion is a paid API for AI agents from agentrxi.com, paid per call via x402, $0.04/call, status unknown (last checked 2026-10-15).

Submits a question to three independent AI models (GPT-OSS-120B, Llama 3.3 70B, Gemma 4) and returns their consensus, disagreements, a confidence level, and a synthesized best answer.

## Facts

- Endpoint: GET https://agentrxi.com/v1/second-opinion?utm_source=zero.xyz
- Price: $0.04/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-15
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/agentrxi-second-opinion-28010fad
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_LtpK_g5fQXORbc7EI0hC_

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 agentrxi-second-opinion-28010fad
```

Example prompt: Get me a second opinion on this question using multiple AI models: 'Is it safe to take ibuprofen with blood pressure medication?' — I want to see where the models agree or disagree and how confident they are.

## When to prefer this

Choose this endpoint when answer reliability and confidence matter — especially for medical, legal, financial, or technical questions where a single LLM's hallucination risk is unacceptable. It is superior to calling a single model when you need to surface disagreements, quantify uncertainty, or justify a decision with multi-model consensus. Prefer over single-model endpoints when the cost of a wrong answer outweighs the $0.04 per-call price.

## Known failure modes

- Missing required 'question' parameter returns a 400 or validation error
- One or more underlying models may time out, potentially reducing consensus quality
- Ambiguous or very short questions may produce low confidence scores
- Payment failure (x402) if USDC balance is insufficient
- Async mode may return a job ID rather than immediate results

## How this service works

Second opinion: three different AI models (gpt-oss-120b, Llama 3.3 70B, Gemma 4) answer independently, then you get the consensus, the points they disagree on, a confidence level and a combined best answer. Also callable as GET with the same fields as query parameters.

## Output

Returns the answers from all three models (GPT-OSS-120B, Llama 3.3 70B, Gemma 4) along with a consensus summary, specific points of disagreement between models, an overall confidence score, and a synthesized best combined answer.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "$schema": "https://json-schema.org/draft/2020-12/schema",
 "required": [
  "input"
 ],
 "properties": {
  "input": {
   "type": "object",
   "required": [
    "type",
    "method"
   ],
   "properties": {
    "type": {
     "type": "string",
     "const": "http"
    },
    "method": {
     "enum": [
      "GET"
     ],
     "type": "string"
    },
    "queryParams": {
     "type": "object",
     "required": [
      "question"
     ],
     "properties": {
      "async": {
       "type": "boolean"
      },
      "context": {
       "type": "string"
      },
      "question": {
       "type": "string"
      }
     }
    }
   },
   "additionalProperties": false
  },
  "output": {
   "type": "object",
   "required": [
    "type"
   ],
   "properties": {
    "type": {
     "type": "string"
    },
    "example": {
     "type": "object"
    }
   }
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/agentrxi-second-opinion-28010fad/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from agentrxi.com](https://www.zero.xyz/host/agentrxi.com/llms.txt)