BenchAtlas

Rankings / xAI

Grok 3 Beta

released 2025-04-09

BenchAtlas Index

as of 2026-07-14
Not enough benchmark coverage for a composite score (needs ≥4 families across ≥3 categories). The per-benchmark evidence below stands on its own — sparse coverage is an evidence gap, not a low score.

Benchmark evidence

6 results
external indices
ECI139.0 pointsindependentEpoch AI Benchmarking Hub
knowledge science
GPQA Diamond67.6 %independentEpoch AI Benchmarking Hub
2025-05-26
reasoning math
FrontierMath3.8 %independentEpoch AI Benchmarking Hub
2025-04-10
FrontierMath Tier 4older version0.0 %independentEpoch AI Benchmarking Hub
2025-07-01
MATH Level 588.8 %independentEpoch AI Benchmarking Hub
2025-04-10
OTIS Mock AIME 2024–202555.6 %independentEpoch AI Benchmarking Hub
2025-04-10

Agent + model results

systems, not bare-model scores
agent + model terminus-1 + Grok 3 BetaTerminal-Bench 1.017.5 %communityTerminal-Bench Leaderboard

These scores measure the whole agent system (scaffold, tools, budgets) — they are never merged into the bare model’s numbers.