Grok 4 benchmark ranking on every AI leaderboard
closed weights · #48 overall, podium score 2.0 · top 10 on 2 of 8 · released Jul 9, 2025
| Rank | Board | Score | |||
|---|---|---|---|---|---|
| 8 |  BFCLAgents and tools · as “Grok-4-0709 (Prompt)” | Agents and tools | 63.0% | 81 | |
| 28 |  Search ArenaAgents and tools · as “grok-4-search” | Agents and tools | 1,142 | 33 | |
| 41 |  ARC-AGI-2Reasoning · as “Grok 4 (Thinking)” | Reasoning | 16.0% | 78 | |
| 49 |  GPQA DiamondReasoning · as “grok-4-0709” | Reasoning | 87.0% | 100 | |
| 68 | Epoch ECIReasoning · as “Grok 4” | Reasoning | 146.5 | 100 | |
| 75 | Vision ArenaVision · as “grok-4-0709” | Vision | 1,183 | 100 | |
| 5 |  Aider PolyglotCoding · stale since 2025-10-03 · as “grok-4 (high)” | Coding | 79.6% | 50 | |
| 18 |  MMLU-ProReasoning · stale since 2026-03-11 · as “Grok-4” | Reasoning | 87.0% | 100 |
Category ranks and more models
Sources: each leaderboard's own published results page, captured every day at 04:30 UTC. Scores exactly as each board published them; we run no evaluations. Logos via logo.dev; trademarks belong to their owners.