# Spraay GPU/Compute — AI Model Inference via Replicate

> Spraay GPU/Compute — AI Model Inference via Replicate is a paid API for AI agents from gateway.spraay.app, paid per call via x402, $0.06/call, status unknown (last checked 2026-09-16).

Runs AI model inference (image, video, LLM, audio, utility) on Replicate's platform via the Spraay gateway, billed per call in USDC.

## Facts

- Endpoint: POST https://gateway.spraay.app/api/v1/gpu/run
- Price: $0.06/call
- Payment: x402
- Status: unknown
- Last checked: 2026-09-16
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/spraay-gpu-compute-ai-model-inference-via-replicate-b1417807
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_sdKKVhYhyzhqHH8o0mGea

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 spraay-gpu-compute-ai-model-inference-via-replicate-b1417807 -d '<json body>'
```

Example prompt: Run the Stable Diffusion XL model on Replicate with the prompt 'a photorealistic mountain landscape at sunrise, 4K' and return the generated image.

## When to prefer this

Choose this endpoint when you need to run any model hosted on Replicate (image, video, audio, LLM, or utility) and want a simple pay-per-call USDC billing model without managing Replicate API keys or billing accounts directly. Ideal for agents that need GPU inference on demand with crypto-native payment via x402.

## Known failure modes

- Invalid or unknown model name/version returns a 400 or model-not-found error
- Malformed input object mismatched to the model's expected schema causes inference failure
- Model cold-start may cause timeouts for rarely-used models
- Payment failure (insufficient USDC balance) returns a 402 error
- Webhook URL unreachable results in notification delivery failure
- Replicate upstream outage causes downstream inference failure

## How this service works

GPU/Compute — run AI model inference via Replicate (image, video, LLM, audio, utility).

## Output

Returns the model's inference output, which varies by model type: image URLs for image generation models, text strings for LLMs, audio file URLs for audio models, or video URLs for video generation models. The exact response structure mirrors Replicate's prediction output schema.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "input": {
   "type": "object"
  },
  "model": {
   "type": "string"
  },
  "version": {
   "type": "string"
  },
  "webhook": {
   "type": "string"
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/spraay-gpu-compute-ai-model-inference-via-replicate-b1417807/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from gateway.spraay.app](https://www.zero.xyz/host/gateway.spraay.app/llms.txt)