Grok 4.7vsGrok 4.5
Across 5 shared benchmarks, Grok 4.7 leads overall: Grok 4.7 wins 4, Grok 4.5 wins 1, with 0 ties and an average score difference of +121.80.
Grok 4.74 wins(80%)(20%)1 winGrok 4.5
Benchmark scores
Grouped by capability, sorted by largest gap within each. 5 shared benchmarks.
Productivity Knowledge
Grok 4.7 2/3| Benchmark | Grok 4.7 | Grok 4.5 | Diff |
|---|---|---|---|
| AA-Briefcase | 1,6572 / 85Thinking High (With Tools) | 1,28337 / 85Thinking High (With Tools) | +374 |
| GDPval-AA v2 | 1,6956 / 107Thinking High (With Tools) | 1,43042 / 107Thinking High (With Tools) | +265 |
| Harvey Lab-AA | 19.6042 / 44Thinking High (With Tools) | 92.429 / 44Thinking High (With Tools) | -72.82 |
AI Agent - Tool Usage
Grok 4.7 1/1| Benchmark | Grok 4.7 | Grok 4.5 | Diff |
|---|---|---|---|
| Terminal-Bench 4.0 | 3816 / 91Thinking High (With Tools) | 12.4248 / 91Thinking High (With Tools) | +25.58 |
Coding and Software Engineer
Grok 4.7 1/1| Benchmark | Grok 4.7 | Grok 4.5 | Diff |
|---|---|---|---|
| DeepSWE | 7116 / 89Thinking High (With Tools) | 53.7657 / 89Thinking High (With Tools) | +17.24 |
Specs
| Field | Grok 4.7 | Grok 4.5 |
|---|---|---|
| Publisher | xAI | xAI |
| Release date | 2026-09-21 | 2026-07-08 |
| Model type | Coding model | Coding model |
| Architecture | Dense | Dense |
| Parameters | Not available | Not available |
| Context length | 500K | 500K |
| Max output | Not available | Not available |
API pricing
Prices use DataLearner records when available; missing fields are not inferred.
| Item | Grok 4.7 | Grok 4.5 |
|---|---|---|
| Text input | $2 / 1M tokens | $2 / 1M tokens |
| Text output | $6 / 1M tokens | $6 / 1M tokens |
| Cache read | $0.5 / 1M tokens | $0.5 / 1M tokens |
Summary
- Grok 4.7leads in:Productivity Knowledge (2/3), AI Agent - Tool Usage (1/1), Coding and Software Engineer (1/1)
On average across the 5 shared benchmarks, Grok 4.7 scores 121.80 higher.
Largest single-benchmark gap: AA-Briefcase — Grok 4.7 1,657 vs Grok 4.5 1,283 (+374).
Page generated from structured model, pricing and benchmark records. No real-time LLM is used to write the prose.