Kottos AI Observatory
Compare LLM list prices by provider venue and tier, plus live time-to-first-token probes
Answered our latest tool-list check 22 hours ago with 2 tools. It will be marked verified after 3 successful checks in 7 days.
- Pricing
- Free
- Platforms
- API, Web
- Runs
- Remotely
- Version
- 1.0.2
- Listing updated
- 1 day ago
About Kottos AI Observatory
Kottos AI Observatory is a free MCP server for ai models and llm tooling. It runs as a hosted remote server, so there is nothing to install locally.
Where the same model is sold by several providers, price and speed differ. This server answers from two tools that need no sign-in: list_models gives list prices for each venue and service tier that sells a given model, and compare_venues sets venues side by side including time to first token. It is the public, anonymous part of Kottos AI and is listed in Claude's connector directory under the Observatory name.
The latency figure comes from a probe that sends one short, fixed prompt to each measured offer once a minute from Ashburn, Virginia, through the Kottos AI gateway. The site cautions that this number is mostly network travel plus the provider's queue, which is less than a long prompt would wait, and that not every offer is probed, which the tool says. Prices are list prices collected from providers' public pages and may be out of date or wrong, and your contract can differ.
The account server at a separate address adds sign-in, usage and savings tools. The site puts a limit of 120 requests a minute from one address and says it counts requests by client name, method, tool and matched model without storing IP addresses.
Install
Run in a terminal
claude mcp add --transport http venues https://data.kottos.ai/mcpAdd to Settings → Connectors → Add custom connector
https://data.kottos.ai/mcpAdd to ~/.cursor/mcp.json
{
"mcpServers": {
"venues": {
"url": "https://data.kottos.ai/mcp"
}
}
}Add to .vscode/mcp.json
{
"servers": {
"venues": {
"type": "http",
"url": "https://data.kottos.ai/mcp"
}
}
}Add to ~/.codeium/windsurf/mcp_config.json
{
"mcpServers": {
"venues": {
"serverUrl": "https://data.kottos.ai/mcp"
}
}
}Add to your client's MCP settings (most use this shape)
{
"mcpServers": {
"venues": {
"type": "http",
"url": "https://data.kottos.ai/mcp"
}
}
}Tools this server provides (2)
Read directly from the server.
- list_models
- Every model Kottos AI lists prices for, with how many venue and service tier offers sell it and how many of them the live probe measures. Its names are the ones compare_venues accepts.
- compare_venues
- For one model, each venue and service tier that sells it: list prices in US dollars per million tokens and, where the probe measures it, time to first token over the last hour. The probe sends a short prompt once a minute from Ashburn, Virginia, through the Kottos AI gateway, so the time is mostly round trip and queue, not a long prompt's prefill. Measured offers come first, fastest first.
What it does
- Venue price lists
- list_models shows list prices for each venue and service tier selling a model.
- Venue comparison
- compare_venues lines up venues for a model by price and latency.
- Minute-by-minute probe
- Time to first token is measured once a minute from Ashburn, Virginia.
- No sign-in
- Price and probe tools run at data.kottos.ai/mcp without an account.
- Stated caveats
- The tool says which offers are not probed, and prices are labelled as list prices that may be stale.
Good for
- Find which provider sells a model cheapest
- Compare service tiers for the same model
- Check which venue responds fastest right now
- Sanity-check list prices before choosing a provider
- Add live model pricing context to an agent
Pricing
Free
Price and probe tools need no account
Prices change. Check kottos.ai before you buy.
Kottos AI Observatory FAQ
Do I need an account?
No. The price and probe tools run with no sign-in.
How are prices gathered?
They are list prices collected from providers' public pages and can be out of date or wrong.
How is latency measured?
A short fixed prompt is sent to each measured offer once a minute from Ashburn, Virginia, through the gateway.
Is every offer probed?
No. The tool says which offers are measured.
Is there a rate limit?
Yes, 120 requests a minute from one address.
Is Kottos AI Observatory free?
Yes. Kottos AI Observatory is free to use.
How do I add Kottos AI Observatory to Claude Code?
Run this command in your terminal: claude mcp add --transport http venues https://data.kottos.ai/mcp
What are the best alternatives to Kottos AI Observatory?
The closest alternatives to Kottos AI Observatory are Claude Code Ultimate Guide, signafAIde, Toolkien, Lugh and HypePaper.
Alternatives to Kottos AI Observatory

Claude Code Ultimate Guide
MCP serverMCP server for searching the Claude Code Ultimate Guide from any MCP client
Runs locally- Open source
- 6.5K GitHub stars
- AI models and LLM tooling

signafAIde
MCP serverLocal MCP server to set up, verify and check coverage of signed agent action records
Runs locally- 5 tools
- Freemium
- Security

Toolkien
MCP serverOne MCP gateway for all your MCP servers, with code mode, permissions and audit
Not checked yet- 3 tools
- Remote
- Open source
- AI models and LLM tooling
Lugh
MCP serverWrite, check and dry-run Lugh agent programs with policy, budgets and approvals
Runs locally- 5 tools
- Open source
- AI models and LLM tooling
HypePaper
MCP serverResearch papers, benchmarks, citations and topic experts for AI assistants
- 58 tools
- Remote
- Freemium
- AI models and LLM tooling

One compact entry point that finds, describes and runs hundreds of pay-per-call data tools
- 3 tools
- Remote
- Freemium
- Developer tools