Grok 4.5 vs Grok 4.7
Grok 4.7 leads Grok 4.5 by 5.4 points on the Shogo Agent Index, but their ranges overlap, so treat it as close. Grok 4.5 wins 5 and Grok 4.7 wins 13 of 18 shared benchmarks. Grok 4.5 is cheaper for agent workloads.
Add more models to this comparison →Pillars
| Pillar | Grok 4.5 | Grok 4.7 |
|---|---|---|
| Agentic coding | 54 | 58 |
| Long-horizon autonomy | 59 | 69 |
| Knowledge-work agents | 42 | 47 |
| Computer use | 36 | 39 |
| Security & context learning | 48 | 50 |
Shared benchmarks
| Benchmark | Grok 4.5 | Grok 4.7 |
|---|---|---|
| APEX-Agentsindex | 56.2% | 70.0% |
| Vending-Bench 2index | $3,887 | $10,537 |
| ARC-AGI | 87.2% | 90.2% |
| ARC-AGI-2 | 52.6% | 61.4% |
| Chess Puzzles | 35.0% | 41.0% |
| CritPt | 15.4% | 18.0% |
| DTBench | 96.5% | 96.0% |
| FrontierCode | 42.4% | 47.6% |
| FrontierMath Tier 4 | 24.4% | 17.1% |
| FrontierMath Tiers 1-3 | 57.2% | 53.0% |
| GPQA Diamond | 93.4% | 92.7% |
| LMCA | 45.2% | 49.4% |
| MirrorCode | 32.2% | 54.4% |
| OTIS Mock AIME | 97.8% | 98.1% |
| ProofBench | 31.0% | 26.0% |
| SciCode | 55.0% | 57.8% |
| SimpleQA Verified | 48.3% | 56.0% |
| WebDev Arena | 1555 | 1638 |
Specs and price
| Grok 4.5 | Grok 4.7 | |
|---|---|---|
| Agent Index | 49.4 (#36) | 54.8 (#27) |
| Agent $/M | $0.81 | $0.98 |
| Input / output $/M | $2.00 / $6.00 | $2.00 / $6.00 |
| Context | 500K | 500K |
| Released | Jul 8, 2026 | Sep 21, 2026 |
| Weights | Closed | Closed |
| API hosts | 4 | 5 |
Don’t pick one. Route.
Shogo can run Grok 4.5 and Grok 4.7 side by side and send each step of an agent to the one that’s cheaper for it.
Try both in Shogo