AI & Machine Learning
cloudprice.net
An MCP server for searching, comparing, and estimating costs for AI models across providers from a large catalog.
ENDPOINT 1
https://cloudprice.net/api/mcp/data-free
MCP server metadata
- Name
- cloudprice
- Version
- 1.0.0
CloudPrice gives live pricing data for AI models and cloud compute. AI model tools cover 1000+ models (OpenAI, Anthropic, Google, AWS Bedrock, Azure, Mistral, Meta, DeepSeek, Qwen and dozens of inference providers). search_ai_models finds models and doubles as a leaderboard (sort by a benchmark rank, include benchmarks); get_ai_model returns one model with, on request, its full price matrix and benchmark scores; calculate_ai_cost and compare_ai_models price a workload; find_similar_ai_models suggests alternatives; find_gpu_fit matches open-weights models and GPU instances either way. Model ids: pass a CloudPrice canonical id (openai-gpt-5-5, anthropic-claude-4-7-opus), a provider or LiteLLM id (gpt-5.5, claude-opus-4-7) or a fuzzy name; when a tool answers with multiple_matches, retry with one of the listed candidate ids. Prices are per 1M tokens unless the price dimension says otherwise, in the requested currency (USD by default). Cloud tools cover Azure, AWS and GCP: search_cloud_instances lists and filters instance types in one region by vCPU, memory, GPU and price; get_cloud_instance returns one type with specs and per-region prices; find_similar_cloud_instances suggests alternatives. There are two CloudPrice MCP servers with the same tools. On the free one (/api/mcp/data-free) they show the site's free tier: on-demand and spot prices, the first regions, same-cloud alternatives; a call that asks for more returns a sign_in_required error, and the user has to add the subscriber server instead. On the subscriber server (/api/mcp/data, the user signs in to CloudPrice when adding it) a user with a subscription gets everything: reserved and savings-plan prices, all regions (get_cloud_instance full_data), cross-cloud alternatives and more results.
Known tools 12
search_ai_modelsSearch and filter the AI model catalog (1000+ LLMs, embedding, image, audio and video models).
Inferred read-onlyget_ai_modelOne AI model by any id or name: specs, capabilities, context window, ids on every provider.
Inferred read-onlycalculate_ai_costEstimate what a workload costs on one or many models from token counts (input, output, reasoning, cache read/write), images and audio, multiplied by calls.
Potential side effectscompare_ai_modelsSide-by-side prices for 2 to 5 models on one tier: per model, the price of each dimension on every provider that hosts it.
Inferred read-onlyfind_similar_ai_modelsModels similar to a given one (same type, comparable context window and price/performance), ranked by the chosen strategy.
Inferred read-onlyget_cloud_instanceSpecs and prices for one Azure VM size, AWS EC2 instance type or GCP Compute machine type: vCPU, memory, GPU, network, plus per-region prices for the chosen OS and payment type.
Potential side effectssearch_cloud_instancesInstance types available in one Azure, AWS or GCP region with vCPU, memory, GPU and the hourly price for the chosen OS and payment type; filter by size, GPU or name and sort by price.
Potential side effectsfind_similar_cloud_instancesAlternatives to an instance type with comparable vCPU, memory and GPU, with their prices in the same region: cheaper or newer generations in the same cloud, or (signed-in subscriber) the closest match on another cloud.
Inferred read-onlyget_cloud_price_historyHow the price of one instance type moved over time in up to three regions: on-demand list price, spot market price, or (signed-in subscriber) reserved / commitment prices.
Inferred read-onlyget_cloud_regionOne Azure, AWS or GCP region: location, average on-demand price and its rank among the cloud's regions, how many instance types it offers, nearby regions on every cloud with distance, AWS availability zones and service health, and whether a given instance type is offered there.
Inferred read-onlycompare_cloud_instancesOne table for up to five instance types, any mix of Azure, AWS and GCP: vCPU, memory, GPU, architecture and the hourly and monthly price in the chosen region on one price basis (on-demand, spot, or for a signed-in subscriber reserved / 1-year and 3-year commitments), cheapest first.
Inferred read-onlyCONNECT WITH APPROVAL
Client installation
Review this server and its permissions before adding it. Secret placeholders must be set locally.
Codex
~/.codex/config.toml
[mcp_servers.cloudprice]
url = "https://cloudprice.net/api/mcp/data-free"
enabled = true
Claude Code
.mcp.json
{
"mcpServers": {
"cloudprice": {
"type": "http",
"url": "https://cloudprice.net/api/mcp/data-free"
}
}
}
Claude Desktop
Settings → Connectors → Add custom connector
Name: cloudprice
Remote MCP URL: https://cloudprice.net/api/mcp/data-free
Add this remote URL as a custom connector in Claude Desktop. Availability depends on the user plan and workspace policy.
Cursor
.cursor/mcp.json
{
"mcpServers": {
"cloudprice": {
"url": "https://cloudprice.net/api/mcp/data-free"
}
}
}
Visual Studio Code
.vscode/mcp.json
Add to Visual Studio Code{
"servers": {
"cloudprice": {
"type": "http",
"url": "https://cloudprice.net/api/mcp/data-free"
}
}
}
Generic MCP
Client-specific MCP configuration
{
"name": "cloudprice",
"transport": "streamable-http",
"url": "https://cloudprice.net/api/mcp/data-free"
}
MCP Inspector
Run the official MCP Inspector locally and enter the indexed Streamable HTTP endpoint.
ENDPOINT 2
https://cloudprice.net/api/mcp/data
Known tools 0
No tool metadata was available in the registry cache.
CONNECT WITH APPROVAL
Client installation
Review this server and its permissions before adding it. Secret placeholders must be set locally.
Codex
~/.codex/config.toml
[mcp_servers.cloudprice-net]
url = "https://cloudprice.net/api/mcp/data"
enabled = true
bearer_token_env_var = "MCP_BEARER_TOKEN"
Authentication is required. Replace the placeholder locally and never commit a secret.
Claude Code
.mcp.json
{
"mcpServers": {
"cloudprice-net": {
"type": "http",
"url": "https://cloudprice.net/api/mcp/data",
"headers": {
"Authorization": "Bearer YOUR_BEARER_TOKEN"
}
}
}
}
Authentication is required. Replace the placeholder locally and never commit a secret.
Claude Desktop
Settings → Connectors → Add custom connector
Name: cloudprice-net
Remote MCP URL: https://cloudprice.net/api/mcp/data
Add the URL as a custom connector, then complete its supported authorization flow. Claude Desktop remote connectors are configured in the UI.
Cursor
.cursor/mcp.json
{
"mcpServers": {
"cloudprice-net": {
"url": "https://cloudprice.net/api/mcp/data",
"headers": {
"Authorization": "Bearer YOUR_BEARER_TOKEN"
}
}
}
}
Authentication is required. Replace the placeholder locally and never commit a secret.
Visual Studio Code
.vscode/mcp.json
{
"servers": {
"cloudprice-net": {
"type": "http",
"url": "https://cloudprice.net/api/mcp/data",
"headers": {
"Authorization": "Bearer ${input:mcp-token}"
}
}
},
"inputs": [
{
"type": "promptString",
"id": "mcp-token",
"description": "cloudprice-net bearer token",
"password": true
}
]
}
Authentication is required. Replace the placeholder locally and never commit a secret.
Generic MCP
Client-specific MCP configuration
{
"name": "cloudprice-net",
"transport": "streamable-http",
"url": "https://cloudprice.net/api/mcp/data",
"headers": {
"Authorization": "Bearer YOUR_BEARER_TOKEN"
}
}
Authentication is required. Replace the placeholder locally and never commit a secret.
MCP Inspector
Run the official MCP Inspector locally and enter the indexed Streamable HTTP endpoint.
TRUST AND VERIFICATION EVIDENCE
Loading Trust v2 evidence…
Checking the associated registrable domain. The BuiltWith key remains server-side.
Evidence is source-attributed and does not guarantee that a third-party server is safe. Risk labels are conservative metadata heuristics.