Tokeven Cost Advisor
Price a Claude, GPT, Gemini or Grok call before sending it, with local cost estimates
- Pricing
- Freemium
- Platforms
- CLI, API
- Runs
- Locally
- Version
- 0.18.0
- Listing updated
- 1 day ago
About Tokeven Cost Advisor
Tokeven Cost Advisor is a freemium MCP server for ai models and llm tooling. It runs locally on your machine and is started by your AI client.
The local package answers one question before a prompt goes out: what will it cost, and what would it cost on another model? It shows the price on the current model, prices on alternatives across Claude, GPT, Gemini and Grok, and a confidence score for each, leaving the decision to the developer. The server runs over stdio from a PyPI package, and the site says prompt content and provider keys never reach Tokeven.
The wider product wraps an existing client in a couple of lines of Python to record token counts, model, latency and cost, then flags waste such as a repeated large system prompt that could be cached, or simple lookups sent to a premium model. A weekly digest and a Proof dashboard trace savings to calls where a team accepted a cheaper option.
The free plan covers one developer with 1,000 tracked calls a month, a cost dashboard, pre-call estimates over MCP and model recommendations. Team is billed as 10 percent of verified savings, so no savings means no fee. The site's savings figures are modelled examples, not customer averages.
Install
Run in a terminal
claude mcp add tokeven -- uvx tokevenAdd to claude_desktop_config.json
{
"mcpServers": {
"tokeven": {
"command": "uvx",
"args": [
"tokeven"
]
}
}
}Add to ~/.cursor/mcp.json
{
"mcpServers": {
"tokeven": {
"command": "uvx",
"args": [
"tokeven"
]
}
}
}Add to .vscode/mcp.json
{
"servers": {
"tokeven": {
"type": "stdio",
"command": "uvx",
"args": [
"tokeven"
]
}
}
}Add to ~/.codeium/windsurf/mcp_config.json
{
"mcpServers": {
"tokeven": {
"command": "uvx",
"args": [
"tokeven"
]
}
}
}Add to your client's MCP settings (most use this shape)
{
"mcpServers": {
"tokeven": {
"command": "uvx",
"args": [
"tokeven"
]
}
}
}What it does
- Pre-call price estimates
- Shows the cost of a prompt on the current model before sending.
- Cross-provider comparison
- Prices alternatives across Claude, GPT, Gemini and Grok with confidence scores.
- Waste detection
- Flags uncached repeated prompts and premium models used for simple tasks.
- Savings proof
- Dashboard traces accepted recommendations to priced calls.
- Weekly digest
- Admins get spend by model and provider with coaching tips.
- Local and metrics-only
- Runs locally and sends only usage metrics, not prompts or keys.
Good for
- Estimate a prompt's cost before committing
- Decide whether a cheaper model fits a simple lookup
- Find system prompts that should be cached
- Audit AI spend per developer and team
Pricing
Freemium
Free plan; Team charges 10% of verified savings
Prices change. Check tokeven.com before you buy.
Tokeven Cost Advisor FAQ
Do prompts leave my machine?
The site says prompt content and your provider key never reach Tokeven; only token counts, model, latency and cost are captured.
How much does it cost?
Free covers one developer and 1,000 tracked calls a month; Team takes 10% of verified savings and nothing if there are none.
Which providers are covered?
Claude, OpenAI, Gemini and Grok.
How are savings verified?
Only accepted recommendations count, priced against token prices in force when each call ran.
Is Tokeven Cost Advisor free?
Tokeven Cost Advisor has a free plan, with paid plans for more usage or features.
How do I add Tokeven Cost Advisor to Claude Code?
Run this command in your terminal: claude mcp add tokeven -- uvx tokeven
What are the best alternatives to Tokeven Cost Advisor?
The closest alternatives to Tokeven Cost Advisor are REA, Codebase Memory, GitHub, Serena MCP: the IDE for your agent and XERJ.
Alternatives to Tokeven Cost Advisor

REA
MCP serverCLI and MCP server that lets coding agents reverse engineer apps and binaries
Runs locally- Open source
- 464K GitHub stars
- Developer tools

Codebase Memory
MCP serverLocal MCP server that builds a code knowledge graph for AI coding agents
Runs locally- Open source
- 48.7K GitHub stars
- Knowledge and memory

GitHub
MCP serverOfficial MCP server connecting AI assistants to GitHub repos, issues and workflows
Not checked yet- Remote
- Open source
- 34.5K GitHub stars
- Developer tools

Serena MCP: the IDE for your agent
MCP serverMCP toolkit giving coding agents symbol-level code retrieval and editing
Runs locally- Open source
- 31.5K GitHub stars
- Developer tools

XERJ
MCP serverLocal search engine for AI agents with cited answers over your code and docs
Runs locally- Open source
- 19.2K GitHub stars
- Knowledge and memory

Ouroboros
MCP serverSpec-first layer that pins requirements and verifies results of AI coding agent work
Runs locally- Open source
- 6.4K GitHub stars
- Developer tools