Enterprise DNAEnterprise DNA

MCP Servers

sachinuppal/modelcostsaver

Offline MCP server that predicts LLM call cost and recommends the cheapest capable model before you call it. No API keys, no network.

Visit site ↗

Our verdict

Not reviewed yet

We have not reviewed sachinuppal/modelcostsaver yet

No one at Enterprise DNA has checked this listing against its sources. The description comes from public listings gathered automatically. Any fact below shows where it came from and when it was last checked.

How we review records

From the listing

About sachinuppal/modelcostsaver

An offline MCP server that predicts the cost of an LLM call and recommends the cheapest capable model before you make the request. It runs locally with no API keys or network access, using npx @workswarm/modelcostsaver.

Best for

Developers building MCP-based workflows who want to minimize LLM call costs without relying on external services.

Use cases

  • Selecting the most cost-effective model for a given prompt before execution
  • Estimating LLM call costs in local or air-gapped development environments
  • Integrating cost-aware model routing into MCP-based toolchains

Notes

0 stars on GitHub. Last updated 2026-06-23. Licensed Apache-2.0.

Pros

  • No API keys or network required, works fully offline
  • Simple to run via npx with zero setup
  • Helps avoid surprise costs by predicting expenses upfront

Cons

  • Zero GitHub stars indicates very early stage or limited adoption
  • Cost predictions may not reflect real-time pricing changes without network updates
  • Only recommends cheapest model, not necessarily the best for accuracy or latency

Key facts

What we have checked

Not checked yet. Dated facts such as prices, releases and repository activity appear here once we have confirmed them from a source.

How to use

How to set it up

As the listing publishes it. We have not run this install.

Install

npx -y @workswarm/modelcostsaver

Tools exposed

  • estimate_cost
  • predict_cost
  • select_optimal_model
  • compare_models
  • optimize_request
  • record_usage

Clients the listing names (not checked here)

Claude Desktop, Claude Code, Cursor, Windsurf, Cline, VS Code

Install

Add sachinuppal/modelcostsaver to your client

Not verified yet

Our lab has not run this server yet. "Config checked" on a client means the recipe matches that client's documented config format; it does not mean we ran it.

Claude Code

Config checked

Run in your terminal (runs on your computer)

claude mcp add --transport stdio sachinuppal-modelcostsaver -- npx -y @workswarm/modelcostsaver@0.1.1

Version 0.1.1 · From the Official MCP registry, read · Format checked against Claude Code's config docs

Claude Desktop

Config checked

Add to claude_desktop_config.json (runs on your computer)

{
  "mcpServers": {
    "sachinuppal-modelcostsaver": {
      "command": "npx",
      "args": [
        "-y",
        "@workswarm/modelcostsaver@0.1.1"
      ]
    }
  }
}

Version 0.1.1 · From the Official MCP registry, read · Format checked against Claude Desktop's config docs

Cursor

Config checked

Add to .cursor/mcp.json (runs on your computer)

{
  "mcpServers": {
    "sachinuppal-modelcostsaver": {
      "command": "npx",
      "args": [
        "-y",
        "@workswarm/modelcostsaver@0.1.1"
      ]
    }
  }
}

Version 0.1.1 · From the Official MCP registry, read · Format checked against Cursor's config docs

VS Code

Config checked

Add to .vscode/mcp.json (runs on your computer)

{
  "servers": {
    "sachinuppal-modelcostsaver": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "@workswarm/modelcostsaver@0.1.1"
      ]
    }
  }
}

Version 0.1.1 · From the Official MCP registry, read · Format checked against VS Code's config docs

Codex

Config checked

Add to ~/.codex/config.toml (runs on your computer)

[mcp_servers.sachinuppal-modelcostsaver]
command = "npx"
args = ["-y", "@workswarm/modelcostsaver@0.1.1"]

Version 0.1.1 · From the Official MCP registry, read · Format checked against Codex's config docs

  • ChatGPT: ChatGPT connects only to remote servers that need no API key.

Recipes come from the server's entry in the official MCP registry. Check the package name before you run it, and keep keys out of shared files.

Works with

What sachinuppal/modelcostsaver connects to

5 clients, each with how we know and when we checked.

Works withTypeHow we knowChecked
Claude CodeClientInstall recipe matches the client's documented formatRuns locally, version 0.1.1Config checkedClaude Code docs
Claude DesktopClientInstall recipe matches the client's documented formatRuns locally, version 0.1.1Config checkedClaude Desktop docs
CursorClientInstall recipe matches the client's documented formatRuns locally, version 0.1.1Config checkedCursor docs
VS CodeClientInstall recipe matches the client's documented formatRuns locally, version 0.1.1Config checkedVS Code docs
CodexClientInstall recipe matches the client's documented formatRuns locally, version 0.1.1Config checkedCodex docs

Alternatives

Alternatives to sachinuppal/modelcostsaver

RecordTypeHow we knowChecked
LiteLLMThis is an alternative to itOpen sourceNamed in the listingNot dated

Compare

Compare sachinuppal/modelcostsaver side by side

Sources

Where this comes from

No sources recorded yet. Each source we check is listed here with the date we checked it.

A source shows what was published when we checked it. It does not certify security, performance or a working installation. Read our method.

Changelog

What changed

No changes recorded yet. Releases, price changes and deprecations appear here, with their dates, once we track them for this record.

Free 27-page guide

Get the free Developer’s Field Guide

A 27-page field guide to the AI coding workflow with Claude. Claude Code, MCP servers, the prompt patterns that work, and what to delegate. Free.

Enter your work email. We send it straight over, plus a few short notes worth knowing. Unsubscribe any time.

No spam. Unsubscribe any time.

← Back to MCP ServersSubmit your own entry →