Marketplace
No threads yet
Live

get_model_task_costs

artificialanalysis.ai

Returns per-model Intelligence and Coding Index scores, task cost breakdowns, timing, latency, TTFT, and token counts from https://artificialanalysis.ai.

Developer ToolsSports
Parameters1
Runs6
Updated6h ago
Degraded6 runs · 83.3% succeeded
30 days agotoday

Playground

GETget_model_task_costsGETsearch_models_by_nameGETlist_model_summariesGETget_gpu_benchmarksGETget_model_providersGETget_model_detailsGETlist_provider_leaderboardGETlist_media_model_familiesGETget_cost_per_task_metrics

Runs on Notte and spends your workspace credits — what comes back is the live response, not a sample.

What is get_model_task_costs?

Returns per-model Intelligence and Coding Index scores, task cost breakdowns, timing, latency, TTFT, and token counts from https://artificialanalysis.ai. It runs on Notte's browser infrastructure and returns structured JSON, so you can call it like any other API — pass models and get the result back without maintaining a scraper.

Response

{
  "count": 0,
  "records": [
    {}
  ],
  "requested_models": [
    "<string>"
  ]
}

Call it

curl --location 'https://us-prod.notte.cc/functions/0cfd4528-4af5-4b55-9b23-07b0c78c0c13/runs/start' \
--header 'x-notte-api-key: YOUR_API_KEY' \
--header 'Content-Type: application/json' \
--header 'Authorization: Bearer YOUR_API_KEY' \
--data '{
    "function_id": "0cfd4528-4af5-4b55-9b23-07b0c78c0c13",
    "variables": {
        "models": ""
    }
}'

Ask an agent

Paste this into Claude Code, Cursor or any MCP client to have an agent call this function for you.

prompt
Use the Anything API MCP server (https://anything.notte.cc/mcp).

Call run(function_id="0cfd4528-4af5-4b55-9b23-07b0c78c0c13") — the "get_model_task_costs" function, which returns per-model Intelligence and Coding Index scores, task cost breakdowns, timing, latency, TTFT, and token counts from https://artificialanalysis.ai..

It takes models (str). Return the result as structured JSON.