gpu_estimate
Estimate GPU compute service costs before execution to budget AI workloads across multiple providers.
Instructions
Estimate the cost of a GPU-Bridge service before running it. No authentication required.
Input Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| service | Yes | Service key (e.g. llm-4090, image-4090) | |
| seconds | No | Estimated runtime in seconds (optional) |
Implementation Reference
- index.js:201-213 (handler)The handler for the `gpu_estimate` tool in the switch block of the `CallToolRequestSchema` request handler.
case "gpu_estimate": { const { service, seconds } = args; const qs = seconds ? `&seconds=${seconds}` : ""; const est = await apiCall(`/catalog/estimate?service=${service}${qs}`, "GET"); if (est.error) { return { content: [{ type: "text", text: `Error: ${est.error}${est.available_services ? ` Available: ${est.available_services.join(", ")}` : ""}` }], isError: true }; } return { content: [{ type: "text", text: `Service: ${est.service} Estimated cost: $${est.estimated_cost_usd} Rate: $${est.price_per_second}/sec ${est.note}` }] }; } - index.js:62-72 (registration)Registration of the `gpu_estimate` tool within the `ListToolsRequestSchema` handler.
name: "gpu_estimate", description: "Estimate the cost of a GPU-Bridge service before running it. No authentication required.", inputSchema: { type: "object", properties: { service: { type: "string", description: "Service key (e.g. llm-4090, image-4090)" }, seconds: { type: "number", description: "Estimated runtime in seconds (optional)" } }, required: ["service"] } }