# FX Reference Rates API – AI Pricing Bulk Dataset

> FX Reference Rates API – AI Pricing Bulk Dataset is a paid API for AI agents from fx-reference-rates.regnotes.workers.dev, paid per call via x402, $0.1/call, status unknown (last checked 2026-10-30).

Returns a single JSON containing all cloud GPU instance prices (AWS + Azure, 10 regions, on-demand and spot) and all LLM API token prices (~3,300 models), refreshed daily.

## Facts

- Endpoint: GET https://fx-reference-rates.regnotes.workers.dev/v1/ai/dataset?utm_source=zero.xyz
- Price: $0.1/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-30
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/fx-reference-rates-api-ai-pricing-bulk-dataset-54875aa3
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_6mtHkawFyUEL6lOpiDBAg

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 fx-reference-rates-api-ai-pricing-bulk-dataset-54875aa3
```

Example prompt: Give me the full bulk dataset of cloud GPU instance prices across AWS and Azure and LLM API token prices for all ~3,300 models in one shot — I need today's daily-refreshed snapshot with per-source timestamps.

## When to prefer this

Choose this endpoint when you need a comprehensive, single-call snapshot of both cloud GPU instance pricing and LLM API token pricing without having to scrape multiple provider pages. Ideal for cost optimization tools, budget planning, benchmarking dashboards, or any agent that needs to compare AI infrastructure costs across providers and regions in bulk. Prefer this over per-model or per-instance lookup endpoints when you need the full dataset rather than a single record.

## Known failure modes

- Payment not received or invalid — endpoint returns 402 requiring 0.1 USDC payment via x402 protocol
- Dataset temporarily unavailable if upstream cloud or LLM provider pricing pages are unreachable during daily refresh
- Response may be large (thousands of rows); network timeouts possible on slow connections
- Stale data if daily refresh job failed — check generated_at timestamp to verify freshness

## How this service works

Bulk download: every cloud GPU instance price row (AWS + Azure, 10 regions, on-demand and spot, with GPU model/count, $/GPU-hour) and every LLM API token price (~3,300 models) in one JSON with per-source retrieval timestamps. One payment, whole dataset, refreshed daily.

## Output

A JSON object with two top-level arrays: 'gpu' (rows with cloud, instance name, GPU model, GPU count, region, on-demand and spot USD-per-GPU-hour) and 'llm' (rows with model name, input/output/cached-input USD per million tokens, context window), plus a 'generated_at' ISO timestamp and per-source retrieval timestamps. Covers AWS and Azure across 10 regions for GPU, and ~3,300 models for LLM pricing.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "properties": {
   "type": "string"
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "json",
 "example": {
  "gpu": [
   {
    "cloud": "azure",
    "instance": "Standard_ND96isr_H100_v5",
    "usd_per_gpu_hour": 12.29
   }
  ],
  "llm": [
   {
    "model": "gpt-5",
    "input_usd_per_mtok": 1.25
   }
  ],
  "generated_at": "2026-10-25T14:38:09Z"
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/fx-reference-rates-api-ai-pricing-bulk-dataset-54875aa3/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from fx-reference-rates.regnotes.workers.dev](https://www.zero.xyz/host/fx-reference-rates.regnotes.workers.dev/llms.txt)