Compare / head-to-head
Gemini 4 Argonvs
Grok 4.7
Gemini 4 Argon leads 8 of 9 shared benchmarks. Grok 4.7 is 1.5x cheaper per token.
Benchmarks from cited public sources; pricing from official pages; status from official provider feeds.
Shared benchmarks
8 – 0
Gemini 4 Argon leads · 1 tied
Cheaper per token
Grok 4.7
1.5x cheaper, input + output
Larger context
—
unknown
Providers
2 providers
Google · xAI
Gemini 4 Argon
Google · released 2026-09-30
textvisionvideo
Quality
Benchmark matrix
9 shared · 3 only Gemini 4 Argon · 21 only Grok 4.7| Benchmark | Gemini 4 Argon | Grok 4.7 | Δ | Edge |
|---|---|---|---|---|
| Reported by both · 9 | ||||
| LMArena Elo | 1525.2 ↗ | 1442.5 ↗ | +82.7 | Gemini 4 Argon |
| APEX-Agents (Mercor) | 82.2% ↗ | 54.6% ↗ | +27.6 pt | Gemini 4 Argon |
| DeepSWE v1.1 | 77.9% ↗ | 71% ↗ | +6.9 pt | Gemini 4 Argon |
| FrontierSWE V2 (Proximal Labs) | 55% ↗ | 29.5% ↗ | +25.5 pt | Gemini 4 Argon |
| LMArena Agent (LMArena) | 0.0757 ↗ | 0.0404 ↗ | 0 | Tie |
| LMArena WebDev (LMArena) | 1679.5 ↗ | 1637.5 ↗ | +42 | Gemini 4 Argon |
| SAGE (Vals AI) | 53.6% ↗ | 40.8% ↗ | +12.8 pt | Gemini 4 Argon |
| Terminal-Bench 4.0 (Vals AI) | 57.6% ↗ | 28.8% ↗ | +28.8 pt | Gemini 4 Argon |
| Vending-Bench 2 (Andon Labs) | 13718.16 ↗ | 10536.83 ↗ | +3181.3 | Gemini 4 Argon |
| Only Gemini 4 Argon reports · 3 | ||||
| AutomationBench | 51.3% ↗ | not reported | — | — |
| CWE-bench v1 | 68% ↗ | not reported | — | — |
| LVBench | 91.7% ↗ | not reported | — | — |
| Only Grok 4.7 reports · 21 | ||||
| GPQA Diamond | not reported | 92.7% ↗epoch run | — | — |
| AA Briefcase v1.1 | not reported | 1657 ↗ | — | — |
| Chess Puzzles (Epoch AI run) | not reported | 38% ↗epoch run | — | — |
| CursorBench 4.0 | not reported | 46.3% ↗ | — | — |
| EEBench | not reported | 64% ↗ | — | — |
| FrontierMath Tier 4 v2 (Epoch AI run) | not reported | 17.1% ↗epoch run | — | — |
| FrontierMath Tiers 1-3 v2 (Epoch AI run) | not reported | 53% ↗epoch run | — | — |
| Furniture Assembly (Epoch AI run) | not reported | 20.8% ↗epoch run | — | — |
| Harvey Legal Agent Benchmark | not reported | 19.6% ↗ | — | — |
| HealthBench Professional | not reported | 56.7% ↗ | — | — |
| LiveBench Agentic Coding (LiveBench) | not reported | 54% ↗ | — | — |
| LiveBench Coding (LiveBench) | not reported | 77.2% ↗ | — | — |
| LiveBench Data Analysis (LiveBench) | not reported | 76.9% ↗ | — | — |
| LiveBench Instruction Following (LiveBench) | not reported | 75.3% ↗ | — | — |
| LiveBench Language (LiveBench) | not reported | 80.1% ↗ | — | — |
| LiveBench Mathematics (LiveBench) | not reported | 95.7% ↗ | — | — |
| LiveBench Reasoning (LiveBench) | not reported | 82.7% ↗ | — | — |
| Mystery Game Puzzles (Epoch AI run) | not reported | 29% ↗epoch run | — | — |
| OTIS Mock AIME 2024-2025 (Epoch AI run) | not reported | 98.1% ↗epoch run | — | — |
| SimpleQA Verified | not reported | 56% ↗epoch run | — | — |
| 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 Gemini 4 Argon 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 | Gemini 4 Argon | Grok 4.7 | Edge |
|---|---|---|---|
| Context window | — | 500K tokens | — |
| Max output | 1M tokens | — | — |
| Input price / 1M | $2 ↗ | $2 ↗ | Tie |
| Output price / 1M | $10 ↗ | $6 ↗ | Grok 4.7 |
| Cached input / 1M | $0.1 ↗ | $0.5 ↗ | Gemini 4 Argon |
| Input + output / 1M Lower is cheaper. List prices; batch, tool and regional fees excluded. | $12 | $8 | Grok 4.7 |
| Modalities | text · vision · video | text · vision | Gemini 4 Argon |
| Released | 2026-09-30 | 2026-09-21 | — |
| Cited benchmark scores | 12 | 30 | — |
Reliability
Provider status
Live from /statusMore matchups
Gemini 4 Argon 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 Gemini 4 Argon and Grok 4.7 on your own prompts with Respan evals, or route to either through one gateway key with automatic failover.