SLXD AI API Pricing
Measured conclusion: Deep-tested. L2 outcome: tools_ok.
Look up, compare and track AI model API prices from an AI assistant. No sign-in needed.
| Author | app.slxd |
|---|---|
| Access | remote |
| Publisher key required | No |
| Measured state | Deep-tested |
| L1 liveness | alive |
| L1 latency | 2599 ms |
| Last measured | 2026-10-24 |
Add to your agent
Copy this into your MCP client config:
{
"mcpServers": {
"SLXD AI API Pricing": {
"url": "https://aiapipricing.slxd.app/mcp/aiapipricing",
"transport": "streamable-http"
}
}
}
Measured tools (10)
search_modelsread-onlySearches the catalogue of AI models and returns their current prices, with a confidence level and the sources backing each price.
Parameters:
limit— integerquery— string — Free text matched against name and id.provider— string — Restrict to one provider slug, e.g. `openai`.
Hints: read-only yes · destructive not declared · idempotent not declared · open-world not declared
get_model_priceread-onlyReturns every current price for one model — input, output, cached reads, batch and priority tiers, context brackets — with the confidence level and which sources agree. By default these are the maker's own prices; pass `channel` to get what a reseller charges for the same model instead. Two dimensions can share a label, so the ones that would collide carry their qualifiers, e.g. 'Web search (context size: high)'.
Parameters:
model(required) — string — Model id as `provider/slug`, e.g. `anthropic/claude-sonnet-4-5`.channel— string — Sales channel to price. Omit for the maker's own price. `direct` is the maker, the rest resell the same model at their own rate; a channel that does not sell this model returns the model with no prices.
Hints: read-only yes · destructive not declared · idempotent not declared · open-world not declared
compare_modelsread-onlyReturns the current prices of several models side by side. Amounts in credits are never converted to money, so a credit-billed model is not comparable to a token-billed one. Two dimensions can share a label, so the ones that would collide carry their qualifiers, e.g. 'Web search (context size: high)'.
Parameters:
models(required) — array
Hints: read-only yes · destructive not declared · idempotent not declared · open-world not declared
price_historyread-onlyReturns a model's price over time. Each point is [date, amount] with the date as YYYY-MM-DD. The series holds only the points where the price CHANGED — prices are step functions, so a stable model is one point, not one per day; when a price changed twice on one day, both points are returned with the same date, in order. One series per price dimension: two dimensions can share a label, so the ones that would collide carry their qualifiers, e.g. 'Web search (context size: high)'.
Parameters:
model(required) — string — Model id as `provider/slug`, e.g. `anthropic/claude-sonnet-4-5`.to— string — ISO date, inclusive.from— string — ISO date, inclusive.
Hints: read-only yes · destructive not declared · idempotent not declared · open-world not declared
recent_changesread-onlyLists recent price increases, decreases, additions and removals across the catalogue, newest first.
Parameters:
days— integerlimit— integer
Hints: read-only yes · destructive not declared · idempotent not declared · open-world not declared
list_providersread-onlyLists every provider in the catalogue and how many models each has. These are the houses that MAKE the models: a reseller (Vertex, Bedrock, Azure, OpenRouter, a subscription plan) is a sales channel, not a provider — ask `get_model_price` with a `channel` for what one of those charges.
Hints: read-only yes · destructive not declared · idempotent not declared · open-world not declared
data_sourcesread-onlyExplains where the prices come from: which sources are consulted, their weight and licence, when each last succeeded, and whether any is currently failing. Use this to judge how much to trust a price.
Hints: read-only yes · destructive not declared · idempotent not declared · open-world not declared
find_toolsread-onlySearches the full catalogue of available tools by keyword and returns the matching tool names with a short description. Use it when no directly-listed tool fits the task — many tools are available but not listed. Then call tool_schema for the one you picked, and run_tool to execute it.
Parameters:
query(required) — string — Space-separated keywords, e.g. 'invoice pdf export'. Matched against tool names and descriptions; any term may match.
Hints: read-only yes · destructive not declared · idempotent yes · open-world not declared
tool_schemaread-onlyReturns the full input schema of a single tool, including which arguments are required. Call it after find_tools and before run_tool.
Parameters:
name(required) — string — Exact tool name, as returned by find_tools.
Hints: read-only yes · destructive not declared · idempotent yes · open-world not declared
run_toolExecutes a tool by name with the given arguments. Use it for tools that are not directly listed — look them up with find_tools and check their arguments with tool_schema first. Prefer a directly-listed tool whenever one fits. If the action is destructive, ask the user to confirm before calling it.
Parameters:
name(required) — string — Exact tool name, as returned by find_tools.arguments— object — Arguments object matching the tool's schema.
Hints: read-only no · destructive not declared · idempotent not declared · open-world not declared
Measured by MCPYP probing. Untested fields are omitted, never estimated. Listing is always free; MCPYP never proxies calls or stores user tokens.