Compare / head-to-head
GPT-6 Lunavs
Grok 4.7
Grok 4.7 leads 14 of 22 shared benchmarks. GPT-6 Luna is 13.3x cheaper per token. GPT-6 Luna has the larger context window (1.1M tokens).
Benchmarks from cited public sources; pricing from official pages; status from official provider feeds.
Shared benchmarks
7 – 14
Grok 4.7 leads · 1 tied
Cheaper per token
GPT-6 Luna
13.3x cheaper, input + output
Larger context
GPT-6 Luna
1.1M tokens
Providers
2 providers
OpenAI · xAI
GPT-6 Luna
OpenAI · released 2026-09-22
textvision
Quality
Benchmark matrix
22 shared · 0 only GPT-6 Luna · 8 only Grok 4.7| Benchmark | GPT-6 Luna | Grok 4.7 | Δ | Edge |
|---|---|---|---|---|
| Reported by both · 22 | ||||
| GPQA Diamond | 90.5% ↗epoch run | 92.7% ↗epoch run | -2.2 pt | Grok 4.7 |
| LMArena Elo | 1442.9 ↗ | 1442.5 ↗ | +0.4 | GPT-6 Luna |
| APEX-Agents (Mercor) | 44.3% ↗ | 54.6% ↗ | -10.3 pt | Grok 4.7 |
| Chess Puzzles (Epoch AI run) | 31% ↗epoch run | 38% ↗epoch run | -7 pt | Grok 4.7 |
| DeepSWE v1.1 | 66.6% ↗ | 71% ↗ | -4.4 pt | Grok 4.7 |
| FrontierMath Tier 4 v2 (Epoch AI run) | 56.1% ↗epoch run | 17.1% ↗epoch run | +39 pt | GPT-6 Luna |
| FrontierMath Tiers 1-3 v2 (Epoch AI run) | 78.9% ↗epoch run | 53% ↗epoch run | +25.9 pt | GPT-6 Luna |
| Furniture Assembly (Epoch AI run) | 44.2% ↗epoch run | 20.8% ↗epoch run | +23.4 pt | GPT-6 Luna |
| LiveBench Agentic Coding (LiveBench) | 51.2% ↗ | 54% ↗ | -2.8 pt | Grok 4.7 |
| LiveBench Coding (LiveBench) | 79% ↗ | 77.2% ↗ | +1.8 pt | GPT-6 Luna |
| LiveBench Data Analysis (LiveBench) | 73.4% ↗ | 76.9% ↗ | -3.5 pt | Grok 4.7 |
| LiveBench Instruction Following (LiveBench) | 55.9% ↗ | 75.3% ↗ | -19.4 pt | Grok 4.7 |
| LiveBench Language (LiveBench) | 73.8% ↗ | 80.1% ↗ | -6.3 pt | Grok 4.7 |
| LiveBench Mathematics (LiveBench) | 89.1% ↗ | 95.7% ↗ | -6.6 pt | Grok 4.7 |
| LiveBench Reasoning (LiveBench) | 81.8% ↗ | 82.7% ↗ | -0.9 pt | Grok 4.7 |
| LMArena Agent (LMArena) | 0.0135 ↗ | 0.0404 ↗ | 0 | Tie |
| LMArena WebDev (LMArena) | 1578.7 ↗ | 1637.5 ↗ | -58.8 | Grok 4.7 |
| Mystery Game Puzzles (Epoch AI run) | 7% ↗epoch run | 29% ↗epoch run | -22 pt | Grok 4.7 |
| OTIS Mock AIME 2024-2025 (Epoch AI run) | 98.9% ↗epoch run | 98.1% ↗epoch run | +0.8 pt | GPT-6 Luna |
| SAGE (Vals AI) | 48.1% ↗ | 40.8% ↗ | +7.3 pt | GPT-6 Luna |
| SimpleQA Verified | 41.4% ↗epoch run | 56% ↗epoch run | -14.6 pt | Grok 4.7 |
| Terminal-Bench 4.0 (Vals AI) | 13.6% ↗ | 28.8% ↗ | -15.2 pt | Grok 4.7 |
| Only GPT-6 Luna reports · 0 | ||||
| GPT-6 Luna reports nothing Grok 4.7 does not. | ||||
| Only Grok 4.7 reports · 8 | ||||
| AA Briefcase v1.1 | not reported | 1657 ↗ | — | — |
| CursorBench 4.0 | not reported | 46.3% ↗ | — | — |
| EEBench | not reported | 64% ↗ | — | — |
| FrontierSWE V2 (Proximal Labs) | not reported | 29.5% ↗ | — | — |
| Harvey Legal Agent Benchmark | not reported | 19.6% ↗ | — | — |
| HealthBench Professional | not reported | 56.7% ↗ | — | — |
| Terminal-Bench 4.0 | not reported | 38% ↗ | — | — |
| Vending-Bench 2 (Andon Labs) | not reported | 10536.83 ↗ | — | — |
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 GPT-6 Luna 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 | GPT-6 Luna | Grok 4.7 | Edge |
|---|---|---|---|
| Context window | 1.1M tokens | 500K tokens | GPT-6 Luna |
| Max output | 128K tokens | — | — |
| Input price / 1M | $0.1 ↗ | $2 ↗ | GPT-6 Luna |
| Output price / 1M | $0.5 ↗ | $6 ↗ | GPT-6 Luna |
| Cached input / 1M | $0.01 ↗ | $0.5 ↗ | GPT-6 Luna |
| Input + output / 1M Lower is cheaper. List prices; batch, tool and regional fees excluded. | $0.6 | $8 | GPT-6 Luna |
| Modalities | text · vision | text · vision | Tie |
| Released | 2026-09-22 | 2026-09-21 | — |
| Cited benchmark scores | 22 | 30 | — |
Reliability
Provider status
Live from /statusMore matchups
GPT-6 Luna 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 GPT-6 Luna and Grok 4.7 on your own prompts with Respan evals, or route to either through one gateway key with automatic failover.