Grok 4.6 vs Grok 4.7
Respan Index by category
Each category is scored 0 to 100, and how far ahead the winner is. Leads under a point are close to even.
| Category | Grok 4.6 | Grok 4.7 |
|---|---|---|
| Coding | 34.0 Grok 4.6 | 42.6 Grok 4.7 |
| Agents | 65.8 Grok 4.6 | 70.4 Grok 4.7 |
| Reasoning | 59.4 Grok 4.6 | 42.7 Grok 4.7 |
| Knowledge & language | 51.1 Grok 4.6 | 55.5 Grok 4.7 |
| Human preference | 56.5 Grok 4.6 | 56.9 Grok 4.7 |
Benchmarks both report
Grok 4.7 leads on 15 of the 26 benchmarks both report. Higher is better. Each score links to where it was published.
| Benchmark | Grok 4.6 | Grok 4.7 |
|---|---|---|
| LMArena Elo | 1457 | 1443 |
| LMArena Vision (LMArena) | 1264 | 1250 |
| LMArena WebDev (LMArena) | 1618 | 1638 |
| LMArena Agent (LMArena) | -0.006 | 0.0148 |
| MirrorCode (Epoch AI) | 41.1% | 54.4% |
| Chess Puzzles (Epoch AI) | 33% | 41% |
| APEX-Agents (Mercor) | 65.3% | 70% |
| FrontierMath Tier 4 v2 (Epoch AI) | 31.7% | 17.1% |
| FrontierMath Tiers 1-3 v2 (Epoch AI) | 66% | 53% |
| FrontierSWE V2 (Proximal Labs) | 25.3% | 29.5% |
| Furniture Assembly (Epoch AI) | 40% | 20.8% |
| GPQA Diamond | 94% | 92.7% |
| LiveBench Agentic Coding (LiveBench) | 57% | 54% |
| LiveBench Coding (LiveBench) | 76.8% | 77.2% |
| LiveBench Data Analysis (LiveBench) | 73.9% | 76.9% |
| LiveBench Instruction Following (LiveBench) | 71.9% | 75.3% |
| LiveBench Language (LiveBench) | 83.7% | 80.1% |
| LiveBench Mathematics (LiveBench) | 92.6% | 95.7% |
| LiveBench Reasoning (LiveBench) | 90.5% | 82.7% |
| Mystery Game Puzzles (Epoch AI) | 34% | 29% |
| OTIS Mock AIME 2024-2025 (Epoch AI) | 99.2% | 98.1% |
| SAGE (Vals AI) | 28.9% | 40.8% |
| SimpleQA Verified | 49.3% | 56% |
| Terminal-Bench 4.0 (Vals AI) | 17.2% | 28.8% |
| Vending-Bench 2 (Andon Labs) | 9047.03 | 10536.83 |
| DeepSWE v1.1 | 65.9% | 71% |
FAQ
Is Grok 4.6 better than Grok 4.7?
Grok 4.6 scores 53.4 on the Respan Index (#32), ahead of Grok 4.7 at 53.0 (#34). Grok 4.7 leads on 15 of the 26 benchmarks both report.
Which is cheaper, Grok 4.6 or Grok 4.7?
Grok 4.6 lists at $2 input and $6 output per 1M tokens; Grok 4.7 at $2 and $6. Both cost about the same per token at list price.
Which has the larger context window?
Both accept 500K tokens of context.