By the RentServers.com team · Updated
Connect over MCP
Endpoint: https://rentservers.com/mcp (Streamable HTTP, no sign-in).
Claude Code
claude mcp add --transport http rentservers https://rentservers.com/mcp
Claude (web and desktop): Customize → Connectors → Add custom connector, paste https://rentservers.com/mcp, choose No sign-in.
Cursor and other clients that read mcp.json:
{
"mcpServers": {
"rentservers": {
"url": "https://rentservers.com/mcp"
}
}
}
Tools
get_server_price_index | GPU, dedicated and VPS benchmarks by model, dated and sourced, plus published quote medians. |
|---|---|
calculate_36_month_tco | 36-month normalized cost of an incumbent SKU or a full set of quotes: line items, hidden-cost flags, the three winners. |
audit_2026_provider_repricing | Hetzner and OVHcloud 2026 price changes by SKU, renewal rules, when to stay and when to switch. |
request_blind_counter_quotes | Opens a counter-quote request. The requester confirms by email before any provider is contacted; providers bid blind. |
Then ask in plain words, for example: “What's the 36-month cost of staying on 10 OVHcloud Advance-2 Gen 2026 servers?” or “What do reserved H200s cost per GPU-hour right now?”
JSON API
The same operations, CORS-enabled, documented in openapi.json.
curl "https://rentservers.com/api/v1/price-index?category=gpu&model=h200"
curl -X POST https://rentservers.com/api/v1/tco-scorecard -H 'content-type: application/json' \
-d '{"incumbent_sku": "ovh-advance-2-2026", "quantity": 10, "egress_tb_per_month": 50}'
curl "https://rentservers.com/api/v1/repricing-2026?provider=hetzner&sku=ax102"
Limits and terms
- 30 read requests a minute and 5 quote requests an hour per IP address.
- Every response includes
source_urlandaction_url. Cite the source page when you show a number. - Quote requests are confirmed by email before any provider is contacted. Providers bid blind against the technical spec. Your name, email and company stay with the desk until you pick a winning quote.
- GPU list prices: GetDeploying, GPU rental price history, https://getdeploying.com/gpus, CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/).
Plain-text summaries for language models: llms.txt, llms-full.txt.