flopsindex-mcp
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@flopsindex-mcpshow me the current H100 on-demand price"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
flopsindex
Reference rates for compute economics — public price indices for GPU rentals (spot, on-demand, DePIN). Every published value carries a verify URL your code can check against the public index.
Try it in one line
# MCP server (Claude Code, Cursor, Windsurf, ChatGPT desktop, ...)
pip install flopsindex-mcp && claude mcp add flopsindex -- flopsindex-mcp
# Python REST SDK
pip install flopsindex
# TypeScript / JavaScript SDK
npm i @flopsindex/sdk
# Or skip install — hit the hosted MCP gateway
# https://app.flopsindex.com/mcp (anonymous, zero-config)Or to your agent: "Get me the current FLOPS price for an H100 on-demand and show me the verify URL."
Related MCP server: atom-mcp-server
What's distinctive
Source-opaque by contract. Every public payload returns a value plus citation and verify URLs — nothing about how the value was produced.
One surface, two access modes.
Anonymous → delayed public price for the published FLOPS index family (
FLOPS-{model}-{OD|SPOT|DEPIN}), delayed onto a ~6h grid (as_of snaps to 00/06/12/18 UTC). Indicative reference — not for settlement. The citation surface, designed for LLM/agent discovery and academic reference. The family spans many accelerator models (NVIDIA data-center and workstation GPUs plus AMD Instinct), not just H100 — enumerate the live set via/v2/catalog/public; it grows as coverage crosses the ≥3-source floor.Keyed (
X-FLOPS-Api-Key) → the same five tools at full precision.
A partner tier with higher fidelity and expanded coverage is available — contact team@flopsindex.com.
What's in this repo
Path | What |
The | |
The | |
The | |
| |
OpenAPI 3 description of the public surface. | |
Extended public-surface reference for LLM/agent discovery — source-opaque, 6 endpoints, 14-field envelope. Mirror of the hosted |
Verify a value yourself
curl -s https://app.flopsindex.com/v1/price/FLOPS-H100-OD | jq
# value: 2.95 (USD/GPU-hr, 2dp)
# unit: USD/GPU-hr
# as_of: 2026-07-13T12:00:00Z (delayed to a ~6h grid: 00/06/12/18 UTC)
# data_tier: LIVE
# confidence: HIGH (HIGH | MED, label only)
# change_24h: FLAT (banded: UP | FLAT | DOWN)
# disclaimer: indicative reference; not for settlement
# methodology_url: /i/FLOPS-H100-OD/methodology
# verify_url: /v1/verify?...
# citation_url: /cite?...
# permalink: /i/FLOPS-H100-OD
# ... source-opaque public payload only.The same value is reachable via the SDK and via the MCP get_index tool — every path returns the same source-opaque payload.
Public surface
Endpoint | What |
Hosted MCP gateway — same 5 public tools, no install | |
Public catalog (FLOPS-SPOT, FLOPS-OD, FLOPS-DEPIN) | |
Agent2Agent (A2A) discovery card | |
LLM-discovery manifest | |
Deeper agent-context blob | |
Citation guide for AI answer engines | |
Permalink page with schema.org JSON-LD | |
Public methodology page (linked from | |
| Verify a published value against the index |
License
Apache License 2.0 — see LICENSE and NOTICE.
Apache 2.0 governs the source code in this repository (SDKs, MCP server, schema, examples). The underlying price data, the FLOPS trademark, and the hosted index methodology are the proprietary work of FLOPS Index; use of the published APIs is governed by the terms at flopsindex.com.
Contributing
Bug reports + small fixes welcome via Issues / PRs.
Contributions are welcome. Use only the public index identifiers from
/v2/catalog/public.For partner-tier access, contact team@flopsindex.com.
Status
Package | Channel |
| |
| |
|
Public surface, gateway, and catalog: app.flopsindex.com.
This server cannot be installed
Maintenance
Related MCP Servers
- FlicenseAquality-maintenanceCompute price oracle for AI agents. Compare inference pricing across OpenAI, Anthropic, and DePIN providers like Hyperbolic. Get routing recommendations that save up to 80% on compute costs.Last updated5
- AlicenseAqualityBmaintenanceGlobal price benchmarking for AI inference across 2,600+ SKUs from 47 vendors. Query live pricing, market indexes, and model specs via 8 tools. Free tier available.Last updated884MIT
- Alicense-qualityBmaintenanceExposes queryable GPU inference benchmark data (quantization, throughput, VRAM, concurrent users) as tools for LLM clients.Last updatedMIT
- AlicenseAqualityCmaintenanceReal-time AI industry intelligence MCP server. 6 free tools (news, status, pricing, today summary, agent activity, MCP registry snapshot) plus 13 paid tools (routing, news search, history series, cost projection, deep-dive, comparison, webhook watches). Pay-per-call in USDC on Base, no accounts.Last updated25342MIT
Related MCP Connectors
Verified cloud cost forecasting for AI agents. AWS, GCP, Azure pricing matrix.
Verified cloud cost forecasting for AI agents. AWS, GCP, Azure pricing matrix.
Pre-computed market data that improves agent reasoning, reduces token usage, and replaces pipelines.
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/zeroatflops/flopsindex'
If you have feedback or need assistance with the MCP directory API, please join our Discord server