History
Human-exam questions: SAT, GRE, Chinese civil service.
This benchmark is indexed in the Benchlist registry. We're working on a deeper guide, request priority coverage → if you'd like this article expanded.
How AGIEval is graded
The canonical run uses temperature 0.0 and max_tokens 512. 8062 problems graded by accuracy. License: MIT.
Common pitfalls when reporting AGIEval
The same number can mean very different things depending on how it was produced. The biggest failure modes specific to this benchmark:
- See /methodology for general benchmark-reporting pitfalls.
Live Benchlist leaderboard
Top attested scores from the Benchlist registry, hydrated client-side from /api/runs.json. Self-reported numbers are de-prioritised, attested results from a real signed transcript always rank above vendor-disclosed ones.
Top scores · AGIEval
Full leaderboard →How to ship a AGIEval score that nobody can challenge
Run AGIEval on Benchlist
Benchlist runs the canonical AGIEval sample set, captures every transcript, builds a Merkle commitment, and signs the result with an Ed25519 attestor key. The score lands at a public verify URL anyone can replay, and you can opt into an Aligned Layer ZK anchor on Ethereum L1.
Hosted runner, POST a job and we email the verify URL when it's done:
curl -X POST https://benchlist.ai/api/v1/run \
-H "Authorization: Bearer $BENCHLIST_KEY" \
-H "Content-Type: application/json" \
-d '{
"service": "anthropic-claude",
"model": "claude-sonnet-4.5",
"benchmark": "agieval",
"runs": 1,
"limit": 50,
"proof_system": "signed",
"inference_api_key": "managed"
}'Self-hosted, install benchlist-runner via pip, point it at your inference key, get a signed run.json:
pip install benchlist-runner
benchlist run agieval --service anthropic-claude --model claude-sonnet-4.5 --limit 50
benchlist publish run.jsonFAQ
What is AGIEval?
How do I verify a AGIEval score?
benchlist run agieval and replay the Ed25519 signature at /verify/