The LLM data API
Benchmark scores, rankings, pricing, and metadata for every model — over REST or MCP.
Free·Unlimited requests
Used by people at
380+
Models tracked
Every major model with benchmark scores, pricing, context length, and provider metadata. Updated within hours of release.
50+
Verified benchmarks
Coding, math, reasoning, vision, and more. Scores are source-verified, not self-reported.
11
MCP tools
Connect any AI agent — Cursor, Claude Code, Windsurf — with one-click OAuth. No key to copy.
Request
curl -H "Authorization: Bearer $API_KEY" \
"https://api.zeroeval.com/stats/v1/models?organization=anthropic&limit=2"Response
{
"models": [
{
"id": "claude-4-sonnet",
"name": "Claude 4 Sonnet",
"organization": { "id": "anthropic", "name": "Anthropic" },
"providers": [{ "provider_name": "Anthropic", "input_price_per_m": 3.0 }],
"top_scores": { "coding": 94.2 }
}
],
"next_cursor": "...",
"total": 382
}/stats/v1/modelsCatalog with metadata, pricing, and category scores/stats/v1/models/{id}Full model detail with every benchmark score/stats/v1/benchmarksAll benchmarks with categories and model counts/stats/v1/scoresScore matrix — filter across models and benchmarks/stats/v1/rankingsTrueSkill rankings by category/stats/v1/updatesRecently added models (1–30 day lookback)Start building
Completely free with no usage limits. Sign in to get your API key.