# Combify Compute Image Captioning & Visual Q&A

> Combify Compute Image Captioning & Visual Q&A is a paid API for AI agents from compute.combify.app, paid per call via x402, $0.008/call, status healthy (last checked 2026-10-12, last successful call 2026-10-11).

Answers questions about an image or describes its contents using Llama 3.2 11B vision model for $0.008 USDC per call

## Facts

- Endpoint: POST https://compute.combify.app/caption?utm_source=zero.xyz
- Price: $0.008/call
- Payment: x402
- Status: healthy
- Last checked: 2026-10-12
- Last successful call: 2026-10-11
- Success rate: 100% of calls made through Zero
- Activations on Zero: 15
- Tags: x402
- Canonical page: https://www.zero.xyz/c/combify-compute-image-captioning-visual-q-a-9b20f7d7
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_kRx6qN3J_YKZvK5vMvuMB

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 combify-compute-image-captioning-visual-q-a-9b20f7d7 -d '<json body>'
```

Example prompt: Can you look at this JPEG image I'm attaching and tell me what's in it — specifically, are there any people visible and what are they doing? Keep the answer under 200 tokens.

## When to prefer this

Use this endpoint when the input is an image and you need a description or answer to a visual question. It uses Llama 3.2 11B vision which is well-suited for general scene understanding, object identification, OCR-light tasks, and chart reading. Prefer the sibling /chat endpoint for purely text-based questions. At $0.008 USDC per call it is cost-effective for single-image analysis; payment only settles on success.

## Known failure modes

- 402 Payment Required if the x402 payment header is missing or invalid
- Image exceeds 4 MB decoded size limit — request rejected
- Unsupported image format (only PNG, JPEG, GIF, WEBP accepted)
- max_tokens out of range (must be 1–512)
- Malformed base64 or invalid data URI in image field
- Model timeout on very complex images or long prompts

## How this service works

Answer questions about an image or describe what it shows (Llama 3.2 11B vision), fixed price $0.008 USDC. Use this when the input is an image; use /chat for text-only questions. POST JSON body {image (required, base64 or data:image/... URI, max 4 MB decoded, PNG/JPEG/GIF/WEBP), prompt (optional question, defaults to describing the image), max_tokens (optional 1-512, default 256)}. Returns {text}. Unpaid calls get 402; paid calls settle only on success.

## Output

Returns a JSON object with a 'text' field containing the model's description of the image or its answer to the provided question, up to max_tokens in length.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "image": {
   "type": "string",
   "description": "Base64-encoded image, max 4 MB decoded"
  },
  "prompt": {
   "type": "string",
   "description": "Question about the image"
  },
  "max_tokens": {
   "type": "integer",
   "description": "Max output tokens, 1 to 512"
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/combify-compute-image-captioning-visual-q-a-9b20f7d7/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from compute.combify.app](https://www.zero.xyz/host/compute.combify.app/llms.txt)