Compare / head-to-head
Grok 4.6vsGrok 4.7
Grok 4.6 leads 12 of 24 shared benchmarks. Both list the same combined token price. Both offer a 500K-token context window.
Benchmarks from cited public sources; pricing from official pages; status from official provider feeds.
Shared benchmarks
12 – 11
Grok 4.6 leads · 1 tied
Cheaper per token
Tie
list price, input + output
Larger context
Tie
both 500K tokens
Providers
xAI
same provider
Grok 4.6
xAI · released 2026-08-12
textvision
Quality
Benchmark matrix
24 shared · 14 only Grok 4.6 · 6 only Grok 4.7| Benchmark | Grok 4.6 | Grok 4.7 | Δ | Edge |
|---|---|---|---|---|
| Reported by both · 24 | ||||
| GPQA Diamond | 94% ↗epoch run | 92.7% ↗epoch run | +1.3 pt | Grok 4.6 |
| LMArena Elo | 1453.8 ↗ | 1442.5 ↗ | +11.3 | Grok 4.6 |
| APEX-Agents (Mercor) | 65.3% ↗ | 54.6% ↗ | +10.7 pt | Grok 4.6 |
| Chess Puzzles (Epoch AI run) | 40% ↗epoch run | 38% ↗epoch run | +2 pt | Grok 4.6 |
| DeepSWE v1.1 | 65.9% ↗ | 71% ↗ | -5.1 pt | Grok 4.7 |
| FrontierMath Tier 4 v2 (Epoch AI run) | 31.7% ↗epoch run | 17.1% ↗epoch run | +14.6 pt | Grok 4.6 |
| FrontierMath Tiers 1-3 v2 (Epoch AI run) | 66% ↗epoch run | 53% ↗epoch run | +13 pt | Grok 4.6 |
| FrontierSWE V2 (Proximal Labs) | 25.3% ↗ | 29.5% ↗ | -4.2 pt | Grok 4.7 |
| Furniture Assembly (Epoch AI run) | 40% ↗epoch run | 20.8% ↗epoch run | +19.2 pt | Grok 4.6 |
| LiveBench Agentic Coding (LiveBench) | 57% ↗ | 54% ↗ | +3 pt | Grok 4.6 |
| LiveBench Coding (LiveBench) | 76.8% ↗ | 77.2% ↗ | -0.4 pt | Grok 4.7 |
| LiveBench Data Analysis (LiveBench) | 73.9% ↗ | 76.9% ↗ | -3 pt | Grok 4.7 |
| LiveBench Instruction Following (LiveBench) | 71.9% ↗ | 75.3% ↗ | -3.4 pt | Grok 4.7 |
| LiveBench Language (LiveBench) | 83.7% ↗ | 80.1% ↗ | +3.6 pt | Grok 4.6 |
| LiveBench Mathematics (LiveBench) | 92.6% ↗ | 95.7% ↗ | -3.1 pt | Grok 4.7 |
| LiveBench Reasoning (LiveBench) | 90.5% ↗ | 82.7% ↗ | +7.8 pt | Grok 4.6 |
| LMArena Agent (LMArena) | 0.0128 ↗ | 0.0404 ↗ | 0 | Tie |
| LMArena WebDev (LMArena) | 1619.5 ↗ | 1637.5 ↗ | -18 | Grok 4.7 |
| Mystery Game Puzzles (Epoch AI run) | 34% ↗epoch run | 29% ↗epoch run | +5 pt | Grok 4.6 |
| OTIS Mock AIME 2024-2025 (Epoch AI run) | 99.2% ↗epoch run | 98.1% ↗epoch run | +1.1 pt | Grok 4.6 |
| SAGE (Vals AI) | 28.9% ↗ | 40.8% ↗ | -11.9 pt | Grok 4.7 |
| SimpleQA Verified | 49.3% ↗epoch run | 56% ↗epoch run | -6.7 pt | Grok 4.7 |
| Terminal-Bench 4.0 (Vals AI) | 17.2% ↗ | 28.8% ↗ | -11.6 pt | Grok 4.7 |
| Vending-Bench 2 (Andon Labs) | 9047.03 ↗ | 10536.83 ↗ | -1489.8 | Grok 4.7 |
| Only Grok 4.6 reports · 14 | ||||
| AA-Briefcase | 1577 ↗ | not reported | — | — |
| APEX-Agents | 57.5% ↗ | not reported | — | — |
| APEX-SWE | 56.4% ↗ | not reported | — | — |
| Artificial Analysis Intelligence Index | 61 ↗ | not reported | — | — |
| CursorBench 3.2 | 69.9% ↗ | not reported | — | — |
| DeepSWE v1.1 (Datacurve) | 67.5% ↗ | not reported | — | — |
| EBR-bench (Epoch AI run) | 30.5% ↗epoch run | not reported | — | — |
| FrontierCode v1.1 (Extended) | 61.3% ↗ | not reported | — | — |
| GDPVal-AA v2 | 1753 ↗ | not reported | — | — |
| Harvey LAB (Vals) | 15.8% ↗ | not reported | — | — |
| LMArena Vision (LMArena) | 1263.5 ↗ | not reported | — | — |
| SimpleBench (SimpleBench) | 75.9% ↗ | not reported | — | — |
| Terminal-Bench 3.0 | 26% ↗ | not reported | — | — |
| WeirdML (Håvard Tveit Ihle) | 67.3% ↗ | not reported | — | — |
| Only Grok 4.7 reports · 6 | ||||
| AA Briefcase v1.1 | not reported | 1657 ↗ | — | — |
| CursorBench 4.0 | not reported | 46.3% ↗ | — | — |
| EEBench | not reported | 64% ↗ | — | — |
| Harvey Legal Agent Benchmark | not reported | 19.6% ↗ | — | — |
| HealthBench Professional | not reported | 56.7% ↗ | — | — |
| Terminal-Bench 4.0 | not reported | 38% ↗ | — | — |
Scores tagged "epoch" or "matharena" are independent runs, used only where the lab has not published its own; ⚠ marks rows MathArena flags as released after the competition. Higher is better on every row. Δ is Grok 4.6 minus Grok 4.7 in the benchmark's own unit. "Not reported" means the lab has not published that figure; it is not a zero. ↗ opens the source.
Specs & pricing
Side by side
Official pricing pages and model cards| Spec | Grok 4.6 | Grok 4.7 | Edge |
|---|---|---|---|
| Context window | 500K tokens | 500K tokens | Tie |
| Max output | — | — | — |
| Input price / 1M | $2 ↗ | $2 ↗ | Tie |
| Output price / 1M | $6 ↗ | $6 ↗ | Tie |
| Cached input / 1M | $0.5 ↗ | $0.5 ↗ | Tie |
| Input + output / 1M Lower is cheaper. List prices; batch, tool and regional fees excluded. | $8 | $8 | Tie |
| Modalities | text · vision | text · vision | Tie |
| Released | 2026-08-12 | 2026-09-21 | — |
| Cited benchmark scores | 43 | 30 | — |
Reliability
xAI status
Live from /statusMore matchups
Grok 4.6 vs …
Models sharing the most benchmarksMore matchups
Grok 4.7 vs …
Models sharing the most benchmarksBuilt by Respan
Which one wins on your data?
Public benchmarks are a starting point. Run Grok 4.6 and Grok 4.7 on your own prompts with Respan evals, or route to either through one gateway key with automatic failover.