# GPU Price Scraper - Vast.ai & RunPod Cloud GPU Rentals (`bhole_krishna/gpu-price-scraper`) Actor

GPU Price Scraper collects live cloud GPU rental prices from Vast.ai and RunPod: hourly cost per GPU, VRAM, reliability and performance per dollar. Covers H100, A100, RTX 4090, L40S and 45+ more GPU models, on-demand and interruptible. No API key or proxy needed.

- **URL**: https://apify.com/bhole_krishna/gpu-price-scraper.md
- **Developed by:** [Krishna Dubey](https://apify.com/bhole_krishna) (community)
- **Categories:** AI, Developer tools
- **Stats:** 2 total users, 1 monthly users, 100.0% runs succeeded, 0 bookmarks
- **User rating**: No ratings yet

## Pricing

from $1.00 / 1,000 results

This Actor is paid per event. You are not charged for the Apify platform usage, but only a fixed price for specific events.

Learn more: https://docs.apify.com/actors/running/actors-in-store.md#pay-per-event

## What's an Apify Actor?

An Actor is a serverless cloud program that runs on the Apify platform. It has two run modes.
In Batch mode, an Actor accepts a well-defined JSON input, performs an action which can take anything from a few seconds to a few hours,
and optionally produces a well-defined JSON output, datasets with results, or files in key-value store.
In Standby mode, an Actor provides a web server which can be used as a website, API, or an MCP server.

Apify vocabulary and the platform model are defined once, in the agent quickstart at https://apify.com/agents.md.

## How to integrate an Actor?

If asked about integration, you help developers integrate Actors into their projects.
You adapt to their stack and deliver integrations that are safe, well-documented, and production-ready.

Do not guess an integration path. Every one of them is in the agent quickstart at https://apify.com/agents.md: the Apify MCP server, Agent Skills with the Apify CLI, the JavaScript and Python clients, the REST API, and the account-free path for an agent with no human to sign in. It also carries the rule on stating cost before the first paid run.

For examples already wired to this Actor's own input schema, see the [API](#api) section below.

Each client library has reference documentation the quickstart does not restate: [JavaScript/TypeScript](https://docs.apify.com/api/client/js/docs.md) (`npm install apify-client`) and [Python](https://docs.apify.com/api/client/python/docs.md) (`pip install apify-client`).

# README

## GPU Price Scraper - Vast.ai & RunPod Cloud GPU Rentals

Track what an hour of cloud GPU compute actually costs. This Actor pulls live rental prices for **H100, A100, RTX 4090, L40S** and 45+ more GPU models from **Vast.ai** and **RunPod**, and returns them as one normalised table, so you can compare the same GPU across providers, regions and price types — on-demand or interruptible.

No login, no API key, no proxy — both providers publish this data openly.

### Why

The same GPU can cost three times more on one provider than another, and prices move daily. There is no single place to see them side by side, so teams either overpay or keep a spreadsheet by hand. This Actor replaces the spreadsheet.

### What you get

One row per offer, with the comparable price (`pricePerGpuHourUsd`) already worked out:

| Field | Meaning |
| --- | --- |
| `gpuModel` | GPU name as the provider spells it |
| `pricePerGpuHourUsd` | **Hourly price per single GPU** — the number to compare |
| `pricePerHourUsd` | Total hourly price of the offer |
| `provider` | `vast.ai` or `runpod` |
| `priceType` | `marketplace`, `on-demand` or `spot` |
| `numGpus`, `vramGbPerGpu` | Machine size |
| `cpuRamGb`, `cpuCores`, `diskSpaceGb` | Host specs (Vast.ai) |
| `internetDownMbps`, `internetUpMbps` | Network speed (Vast.ai) |
| `reliability` | Host reliability 0-1 (Vast.ai) |
| `dlPerf`, `dlPerfPerDollar` | Performance score and value score (Vast.ai) |
| `location` | Where the machine sits |

Rows arrive sorted cheapest first.

### Input

| Field | Type | Description |
| --- | --- | --- |
| `sources` | array | `vast.ai`, `runpod`, or both (default) |
| `gpuModels` | array | Model names like `RTX 4090`, `H100 SXM`. Empty discovers what is on the market right now |
| `onlyCheapestPerModel` | boolean | Keep one cheapest row per provider + model + price type. Use this for a price index |
| `maxPricePerHourUsd` | integer | Drop offers above this price |
| `minVramGb` | integer | Drop GPUs with less memory than this |
| `maxItems` | integer | Row cap (default 500) |

#### Example input

```json
{
    "sources": ["vast.ai", "runpod"],
    "gpuModels": ["RTX 4090", "H100 SXM"],
    "onlyCheapestPerModel": true,
    "maxItems": 100
}
```

#### Example output

```json
{
    "provider": "vast.ai",
    "priceType": "marketplace",
    "gpuModel": "RTX 4090",
    "numGpus": 8,
    "vramGbPerGpu": 24,
    "pricePerHourUsd": 0.9616,
    "pricePerGpuHourUsd": 0.1202,
    "reliability": 0.9834,
    "dlPerfPerDollar": 72.72,
    "location": "Texas, US",
    "scrapedAt": "2026-10-29T09:12:04.118Z"
}
```

### Uses

- **Cost control** — see whether your current provider is still competitive before the next training run
- **Price monitoring** — schedule it daily with `onlyCheapestPerModel` to build a GPU price index over time
- **Capacity hunting** — find who actually has H100s available, and where
- **Value shopping** — sort by `dlPerfPerDollar` instead of raw price

### Notes and limits

- Vast.ai returns at most 64 offers per query, so the Actor queries one GPU model at a time to get real coverage. Listing many models means more requests and a longer run.
- RunPod publishes one price per GPU type rather than per machine, so RunPod rows have no host specs, location or reliability.
- Prices are what the providers advertise at the moment of the run. Marketplace offers can disappear between the run and your booking.
- Spelling matters: `H100 SXM` and `H100 PCIE` are different models to the providers.

### FAQ

**Do I need an account with either provider?**
No. Both endpoints are public.

**How often should I run it?**
Daily is enough for a price index. Marketplace supply shifts within hours, so run it right before booking if you are hunting for a specific machine.

**Can you add more providers?**
Lambda, Together and CoreWeave are the obvious next ones. Open an issue on the Actor if you need a particular provider.

# Actor input Schema

## `sources` (type: `array`):

Which GPU providers to collect prices from.

## `gpuModels` (type: `array`):

Model names exactly as the provider spells them, for example 'RTX 4090' or 'H100 SXM'. Leave empty to discover whatever is on the market right now.

## `onlyCheapestPerModel` (type: `boolean`):

Keep one row per provider, GPU model and price type — the cheapest one. Turn this on for a price index, off to see the whole market.

## `maxPricePerHourUsd` (type: `integer`):

Drop offers above this hourly price. Leave empty for no limit.

## `minVramGb` (type: `integer`):

Drop offers whose GPUs have less memory than this. Leave empty for no limit.

## `maxItems` (type: `integer`):

Stop after saving this many rows. Rows are sorted cheapest first, so a small number still gives you the best offers.

## Actor input object example

```json
{
  "sources": [
    "vast.ai",
    "runpod"
  ],
  "gpuModels": [
    "RTX 4090",
    "H100 SXM"
  ],
  "onlyCheapestPerModel": false,
  "maxItems": 500
}
```

# Actor output Schema

## `results` (type: `string`):

Every collected GPU offer, cheapest first.

## `csv` (type: `string`):

The same rows as a spreadsheet.

# API

You can run this Actor programmatically using our API. Below are code examples in JavaScript, Python, and CLI, as well as the OpenAPI specification and MCP server setup.

## JavaScript example

```javascript
import { ApifyClient } from 'apify-client';

// Initialize the ApifyClient with your Apify API token
// Replace the '<YOUR_API_TOKEN>' with your token
const client = new ApifyClient({
    token: '<YOUR_API_TOKEN>',
});

// Prepare Actor input
const input = {
    "gpuModels": [
        "RTX 4090",
        "H100 SXM"
    ]
};

// Run the Actor and wait for it to finish
const run = await client.actor("bhole_krishna/gpu-price-scraper").call(input);

// Fetch and print Actor results from the run's dataset (if any)
console.log('Results from dataset');
console.log(`💾 Check your data here: https://console.apify.com/storage/datasets/${run.defaultDatasetId}`);
const { items } = await client.dataset(run.defaultDatasetId).listItems();
items.forEach((item) => {
    console.dir(item);
});

// 📚 Want to learn more 📖? Go to → https://docs.apify.com/api/client/js/docs

```

## Python example

```python
from apify_client import ApifyClient

# Initialize the ApifyClient with your Apify API token
# Replace '<YOUR_API_TOKEN>' with your token.
client = ApifyClient("<YOUR_API_TOKEN>")

# Prepare the Actor input
run_input = { "gpuModels": [
        "RTX 4090",
        "H100 SXM",
    ] }

# Run the Actor and wait for it to finish
run = client.actor("bhole_krishna/gpu-price-scraper").call(run_input=run_input)

# Fetch and print Actor results from the run's dataset (if there are any)
print(f"💾 Check your data here: https://console.apify.com/storage/datasets/{run.default_dataset_id}")
for item in client.dataset(run.default_dataset_id).iterate_items():
    print(item)

# 📚 Want to learn more 📖? Go to → https://docs.apify.com/api/client/python/docs/quick-start

```

## CLI example

```bash
echo '{
  "gpuModels": [
    "RTX 4090",
    "H100 SXM"
  ]
}' |
apify call bhole_krishna/gpu-price-scraper --silent --output-dataset

```

## MCP server setup

```json
{
    "mcpServers": {
        "apify": {
            "type": "http",
            "url": "https://mcp.apify.com/?tools=fetch-actor-details,bhole_krishna/gpu-price-scraper"
        }
    }
}
```

The hosted server signs you in with OAuth on first connect, so no API token belongs in this config. Clients without OAuth support can send an `Authorization: Bearer <APIFY_API_TOKEN>` header instead, using a token from API & Integrations in Apify Console (https://console.apify.com/settings/integrations).

## OpenAPI specification

Download the OpenAPI definition: https://api.apify.com/v2/actors/7sY1GZtRBXM0qNa4e/builds/Fyu8B5QcdRgFvtm8f/openapi.json