Grok 4.5 (Medium): benchmark results
Provider: xAI. Released 2026-07-08. Access: API.
Aggregate Rating 1801 ± 23, rank #132 of 2170 rated models, from 20 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| ObviousBench | 100 | Answer pass³ (%) | 96.8 |
| PetriBench - MinimumTokenSteps (Hard) | 66.8 | Exact-Match Accuracy (%) | 88.6 |
| DuelLab GameBench 2 - Reasoning Variants | 60.11 | GameBench2 normalized score (0–100) | 72.2 |
| PetriBench - Deadlock (Hard) | 59 | Exact-Match Accuracy (%) | 71.4 |
| PetriBench - Hard | 48.9 | Exact-Match Accuracy (%) | 71.4 |
| PetriBench - Medium | 65 | Exact-Match Accuracy (%) | 68.6 |
| ARC-AGI-1 | 87.17 | Accuracy (%) | 66.1 |
| PetriBench - L4-Liveness (Hard) | 73.5 | Exact-Match Accuracy (%) | 65.7 |
| ARC-AGI-2 | 52.64 | Accuracy (%) | 62.9 |
| PetriBench - Easy | 83.8 | Exact-Match Accuracy (%) | 62.9 |
| PetriBench - ReachableMarkings (Hard) | 12 | Exact-Match Accuracy (%) | 60 |
| RuneBench | 5.35 | Mean ln(1 + peak XP/min) across sixteen 30-minute skills; hi | 58.2 |
Interactive version: theaggregate.ai/model?slug=grok-4-5-medium · How It Works · Data refreshed daily, snapshot 2026-11-05.