Compute Screener

GPU cloud pricing (what an H100, H200 or B200 costs per hour across providers) and AI model API pricing per 1M tokens, with daily history and compute futures. Free, no sign-in.

View on computescreener.com

Hosted MCP Server

npx add-mcp 'https://computescreener.com/mcp'

Installs into Claude Code, Codex, Cursor and more

Documentation

Connect

Ask Claude, ChatGPT, Grok or any MCP app for compute prices: GPU rental prices, compute futures prices and AI token prices. Free, read-only, no sign-in.

https://computescreener.com/mcp

Claude

Add to Claude

  1. Open Compute Screener in Claude's connector directory (the button above) and select Connect.
  2. Or by its URL: Customize › Connectors › Add custom connector, paste the URL, choose no sign-in, then Add.

ChatGPT

  1. Plugins (chatgpt.com/plugins) › + › Add custom MCP server.
  2. Paste the URL, choose no authentication, then create it.

Grok

  1. grok.com/connectors › New Connector › Custom.
  2. Paste the URL.

DeepSeek

DeepSeek's chat app has no custom connectors yet. DeepSeek Harness can use the URL as a streamable-http MCP server.

Other apps

Any app that supports remote MCP servers, for example Claude Code: claude mcp add --transport http computescreener https://computescreener.com/mcp

Tools

Each tool answers what a page of this site shows, and nothing more.

  • GPU rental and compute futures prices (get_gpu_prices) — USD / GPU-hour with 1D, 1W and 1M changes. Name one GPU, such as H100 or B200, to also get what its page shows: each provider's average price, the posted prices outside the index and each exchange's compute futures price for that GPU.
  • AI token prices (get_token_prices) — USD / 1M tokens per company: its models' posted prices at the mix of tokens agents send (87% cache read, 12% input, 1% output), weighted by usage. Name one company to also get its models' usage and prices and its estimated spend on OpenRouter.
  • Price history (get_price_history) — any GPU or company, 1 to 1000 days back (30 by default), daily (GPU weekly before 2026-10-23), as its chart.
  • A price at any time (get_gpu_price_at) — a GPU's index at a past date, or at a time within the last few days, as its 1D and 5D charts draw it.
  • What moved (get_price_changes) — each GPU's and company's change over 1, 5, 7 or 30 days (the 1D, 5D, 1W and 1M on this site), and each model whose price moved.
  • AI model API prices (get_model_prices) — each model's posted input, cache read and output prices per 1M tokens, its price at the mix, its usage share and its 1M change; name a model for its daily price, as its chart.
  • GPU prices by provider (get_provider_prices) — each provider's posted prices per GPU, and its average in the index standardized to the benchmark contract, as its page.
  • Break-even throughput (get_workload_cost) — the tokens per second a GPU must sell at a model's list price (3 input: 1 output) to cover its rental, and at a throughput the revenue and gross margin, as each GPU's and model's page.
  • Every other GPU (get_gpu_market_prices) — RTX 5090, RTX 4090, RTX 6000 Ada, L40, A40, L4, GH200 and over 60 more, as Every GPU lists them: each one's lowest and median posted price per GPU-hour today, offers, in stock and providers.
  • Check a price (verify_price) — a final value's manifest: every file it was computed from with its SHA-256, and the command that recomputes it with the published verifier.
  • Coverage and freshness (list_indices, get_data_status) — every GPU and company listed, and as the status page shows: when the data was last calculated and each source's status.
  • Search and read (search, fetch) — find and read the site's GPU, company, model, provider, futures and methodology pages, as ChatGPT's deep research does.

Every tool only reads the published prices. Nothing is changed, and nothing about your requests is kept. The answers are CC BY 4.0: reuse them, including commercially, with credit to Compute Screener.

Try asking

  • What does it cost to rent an H100 per hour, and how has that changed this month?
  • Compare the rental prices of H200, B200 and B300.
  • What are H100 compute futures trading at, and on which exchanges?
  • Which AI company's tokens cost the most, and how have prices moved this week?
  • Show OpenAI's token price over the last 90 days.

Troubleshooting

  • No tools appear: check the URL is exactly the one above and that no sign-in is chosen, then remove the connector and add it again.
  • "Not listed yet": a GPU or company is listed once it has a price, a month of history and a value for each of its 1-day, 1-week and 1-month changes.
  • "Prices are unavailable right now": try again in a minute; this website shows the same figures.
  • "Too many requests": one network address can send up to 150 requests in 10 seconds; wait 10 seconds.
  • Anything else: Support.

Methodology · Free API and CSV · Support · Privacy · Terms