Skip to main content
Glama

corpusAI Cloud Pricing

by io.github.JTraversa
OverviewDiscussionsAdmin

GPU term structure

term
Read-onlyIdempotent

Paid (0.05 USDC per call). Term structure per hyperscaler for one GPU model, USD per GPU-hour: spot, on-demand, 1y, 3y. For each hyperscaler with a mapped instance family (see /gpu/models hyperscaler_spot_legs), the median across regions of each term's price divided by GPUs per node, with lo/hi across regions and the discount to on-demand. Answers "what does a committed H100 cost against spot and on-demand, per cloud". Neo-cloud commitment prices are quote-only and excluded.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
gpuYesNormalized GPU model with a hyperscaler mapping, e.g. `H100 SXM`, `A100 SXM 80GB`, `L4`, `T4`. Example: H100 SXM

Output Schema

TableJSON Schema
NameRequiredDescriptionDefault
gpuNo
noteNo
unitNo
providersNo
methodologyNo

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed1 schema field changed
    • changedOutput schema / (root)
      Previous value: -nullNew value: +{
+  "$schema": "http://json-schema.org/draft-07/schema#",
+  "additionalProperties": true,
+  "properties": {
+    "gpu": {
+      "type": "string"
+    },
+    "methodology": {
+      "type": "string"
+    },
+    "note": {
+      "type": "string"
+    },
+    "providers": {
+      "additionalProperties": true,
+      "properties": {},
+      "type": "object"
+    },
+    "unit": {
+      "type": "string"
+    }
+  },
+  "type": "object"
+}
  2. Changed1 schema field changed
    • changedInput schema / properties / gpu / description
      Previous value: -"Normalised GPU model with a hyperscaler mapping, e.g. `H100 SXM`, `A100 SXM 80GB`, `L4`, `T4`. Example: H100 SXM"New value: +"Normalized GPU model with a hyperscaler mapping, e.g. `H100 SXM`, `A100 SXM 80GB`, `L4`, `T4`. Example: H100 SXM"
  3. Changed1 schema field changed
    • changedInput schema / properties / gpu / description
      Previous value: -"Normalized GPU model with a hyperscaler mapping, e.g. `H100 SXM`, `A100 SXM 80GB`, `L4`, `T4`. Example: H100 SXM"New value: +"Normalised GPU model with a hyperscaler mapping, e.g. `H100 SXM`, `A100 SXM 80GB`, `L4`, `T4`. Example: H100 SXM"
  4. Changed1 schema field changed
    • changedInput schema / properties / gpu / description
      Previous value: -"Normalised GPU model with a hyperscaler mapping, e.g. `H100 SXM`, `A100 SXM 80GB`, `L4`, `T4`. Example: H100 SXM"New value: +"Normalized GPU model with a hyperscaler mapping, e.g. `H100 SXM`, `A100 SXM 80GB`, `L4`, `T4`. Example: H100 SXM"
  5. Changed1 schema field changed
    • changedInput schema / properties / gpu / description
      Previous value: -"Normalised GPU model with a hyperscaler mapping, e.g. `H100 SXM`, `A100 SXM 80GB`, `L4`, `T4`."New value: +"Normalised GPU model with a hyperscaler mapping, e.g. `H100 SXM`, `A100 SXM 80GB`, `L4`, `T4`. Example: H100 SXM"
  6. First observed

TDQS

A4.4/5.0
Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already cover the safety profile (readOnly, idempotent, non-destructive, openWorld), and the description adds genuinely new behavioral context: a per-call cost of 0.05 USDC, the aggregation methodology (median across regions, divided by GPUs per node, with lo/hi and discount-to-on-demand), and an explicit data exclusion.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Front-loads the cost and the core output, then methodology, then exclusions — every sentence carries information. The middle sentence is long and comma-heavy, but nothing is filler.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

An output schema exists so return values need no explanation; the description instead supplies the aggregation semantics, the pricing unit, and the excluded data needed to interpret results correctly. Nothing an agent needs to call or interpret this tool is missing.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

With one parameter and 100% schema description coverage, the schema already documents that gpu is a normalized model with a hyperscaler mapping and gives examples. The description adds only that the mapping is a prerequisite via /gpu/models hyperscaler_spot_legs, so baseline 3 is appropriate.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb (returns term structure) and resource (per-hyperscaler GPU pricing) plus the exact term buckets (spot, on-demand, 1y, 3y) and unit (USD per GPU-hour). An agent can distinguish it from siblings like ondemand, reserved, and spot_latest without opening any schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The framing question 'what does a committed H100 cost against spot and on-demand, per cloud' gives clear context for when to reach for this tool, and it names an exclusion (neo-cloud commitment prices are quote-only and excluded). It stops short of explicitly routing to sibling alternatives like term_instance or reserved.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All corpusAI Cloud Pricing toolsInspect
Try in Browser
One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources