Gemini 4 Carbon vs Grok 4.7
| Spec | Gemini 4 Carbon | Grok 4.7 |
|---|---|---|
| Respan Index | 76.1 (#3) | 53.0 (#34) |
| Blended price per 1M | $4 | $3 |
| Context window | 1M | 500K |
| Input / output per 1M | $2 / $10 | $2 / $6 |
| Max output | - | - |
| Input | text, image, audio, video | text, vision |
| Released | 2026-11-10 | 2026-09-21 |
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 | Gemini 4 Carbon | Grok 4.7 |
|---|---|---|
| Coding | 69.3 Gemini 4 Carbon | 42.6 Grok 4.7 |
| Agents | 83.5 Gemini 4 Carbon | 70.4 Grok 4.7 |
| Reasoning | 84.1 Gemini 4 Carbon | 42.7 Grok 4.7 |
| Knowledge & language | 77.8 Gemini 4 Carbon | 55.5 Grok 4.7 |
| Human preference | 60.5 Gemini 4 Carbon | 56.9 Grok 4.7 |
Benchmarks both report
Gemini 4 Carbon leads on 15 of the 15 benchmarks both report. Higher is better. Each score links to where it was published.
| Benchmark | Gemini 4 Carbon | Grok 4.7 |
|---|---|---|
| LMArena WebDev (LMArena) | 1769 | 1639 |
| LMArena Elo | 1527 | 1443 |
| LMArena Agent (LMArena) | 0.161 | 0.0148 |
| APEX-Agents (Mercor) | 84.2% | 70% |
| Chess Puzzles (Epoch AI) | 78% | 41% |
| FrontierMath Tier 4 v2 (Epoch AI) | 95.1% | 17.1% |
| FrontierMath Tiers 1-3 v2 (Epoch AI) | 90.5% | 53% |
| FrontierSWE V2 (Proximal Labs) | 61.2% | 29.5% |
| Furniture Assembly (Epoch AI) | 77.5% | 20.8% |
| GPQA Diamond | 96.5% | 92.7% |
| Mystery Game Puzzles (Epoch AI) | 73% | 29% |
| OTIS Mock AIME 2024-2025 (Epoch AI) | 100% | 98.1% |
| SimpleQA Verified | 77.7% | 56% |
| Terminal-Bench 4.0 (Vals AI) | 63.1% | 28.8% |
| Vending-Bench 2 (Andon Labs) | 16484.21 | 10536.83 |
FAQ
Is Gemini 4 Carbon better than Grok 4.7?
Gemini 4 Carbon scores 76.1 on the Respan Index (#3), ahead of Grok 4.7 at 53.0 (#34). Gemini 4 Carbon leads on 15 of the 15 benchmarks both report.
Which is cheaper, Gemini 4 Carbon or Grok 4.7?
Gemini 4 Carbon lists at $2 input and $10 output per 1M tokens; Grok 4.7 at $2 and $6. Grok 4.7 is 1.3x cheaper per token at list price.
Which has the larger context window?
Gemini 4 Carbon accepts 1M tokens and Grok 4.7 500K tokens.