Compare / head-to-head
Claude Sonnet 5vs
Grok 4
Claude Sonnet 5 leads 10 of 10 shared benchmarks. Claude Sonnet 5 has the larger context window (1M tokens).
Benchmarks from cited public sources; pricing from official pages; status from official provider feeds.
Shared benchmarks
10 – 0
Claude Sonnet 5 leads
Cheaper per token
—
list price, input + output
Larger context
Claude Sonnet 5
1M tokens
Providers
2 providers
Anthropic · xAI
Claude Sonnet 5
Anthropic · released 2026-06-30
textvision
Grok 4
xAI · released 2025-07-09
textvision
- Context
- 256K
- Max out
- —
- Input /1M
- —
- Output /1M
- —
- Cached /1M
- —
- Scores
- 17 · 8 core
Quality
Benchmark matrix
10 shared · 21 only Claude Sonnet 5 · 7 only Grok 4| Benchmark | Claude Sonnet 5 | Grok 4 | Δ | Edge |
|---|---|---|---|---|
| Reported by both · 10 | ||||
| GPQA Diamond | 90.5% ↗epoch run | 87% ↗epoch run | +3.5 pt | Claude Sonnet 5 |
| Humanity's Last Exam (no tools) | 43.2% ↗ | 25.4% ↗ | +17.8 pt | Claude Sonnet 5 |
| LMArena Elo | 1461.8 ↗ | 1410.8 ↗ | +51 | Claude Sonnet 5 |
| Chess Puzzles (Epoch AI run) | 35% ↗epoch run | 28% ↗epoch run | +7 pt | Claude Sonnet 5 |
| Humanity's Last Exam (with tools) | 57.4% ↗ | 38.6% ↗ | +18.8 pt | Claude Sonnet 5 |
| LMArena Vision (LMArena) | 1264.5 ↗ | 1184.2 ↗ | +80.3 | Claude Sonnet 5 |
| OTIS Mock AIME 2024-2025 (Epoch AI run) | 94.7% ↗epoch run | 84% ↗epoch run | +10.7 pt | Claude Sonnet 5 |
| SAGE (Vals AI) | 48.9% ↗ | 25.1% ↗ | +23.8 pt | Claude Sonnet 5 |
| SimpleBench (SimpleBench) | 60.6% ↗ | 60.5% ↗ | +0.1 pt | Claude Sonnet 5 |
| WeirdML (Håvard Tveit Ihle) | 68.8% ↗ | 45.7% ↗ | +23.1 pt | Claude Sonnet 5 |
| Only Claude Sonnet 5 reports · 21 | ||||
| SWE-bench Verified | 85.2% ↗ | not reported | — | — |
| APEX-Agents (Mercor) | 54.5% ↗ | not reported | — | — |
| DeepSWE v1.1 (Datacurve) | 53.8% ↗ | not reported | — | — |
| FrontierMath Tier 4 v2 (Epoch AI run) | 29.3% ↗epoch run | not reported | — | — |
| FrontierMath Tiers 1-3 v2 (Epoch AI run) | 65.6% ↗epoch run | not reported | — | — |
| GSO Opt@1 (GSO) | 36.3% ↗ | not reported | — | — |
| LiveBench Agentic Coding (LiveBench) | 59.4% ↗ | not reported | — | — |
| LiveBench Coding (LiveBench) | 80.7% ↗ | not reported | — | — |
| LiveBench Data Analysis (LiveBench) | 71.7% ↗ | not reported | — | — |
| LiveBench Instruction Following (LiveBench) | 63.9% ↗ | not reported | — | — |
| LiveBench Language (LiveBench) | 75% ↗ | not reported | — | — |
| LiveBench Mathematics (LiveBench) | 92.9% ↗ | not reported | — | — |
| LiveBench Reasoning (LiveBench) | 88.7% ↗ | not reported | — | — |
| LMArena Agent (LMArena) | 0.044 ↗ | not reported | — | — |
| LMArena WebDev (LMArena) | 1539.4 ↗ | not reported | — | — |
| Mystery Game Puzzles (Epoch AI run) | 35% ↗epoch run | not reported | — | — |
| SimpleQA Verified | 33.7% ↗epoch run | not reported | — | — |
| SWE-rebench 2026-05-15 to 2026-07-01 (Nebius) | 56.8% ↗ | not reported | — | — |
| Terminal-Bench 4.0 (Vals AI) | 9.6% ↗ | not reported | — | — |
| Toolathlon-Verified (HKUST) | 71.6% ↗ | not reported | — | — |
| Vending-Bench 2 (Andon Labs) | 6377.7 ↗ | not reported | — | — |
| Only Grok 4 reports · 7 | ||||
| AIME 2025 | not reported | 91.7% ↗ | — | — |
| ARC-AGI-2 | not reported | 15.9% ↗ | — | — |
| BALROG (BALROG) | not reported | 43.6% ↗ | — | — |
| GPQA (no tools) | not reported | 87.5% ↗ | — | — |
| HMMT 2025 (no tools) | not reported | 90% ↗ | — | — |
| LiveCodeBench (Jan - May) | not reported | 79% ↗ | — | — |
| USAMO 2025 (no tools) | not reported | 37.5% ↗ | — | — |
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 Claude Sonnet 5 minus Grok 4 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 | Claude Sonnet 5 | Grok 4 | Edge |
|---|---|---|---|
| Context window | 1M tokens | 256K tokens | Claude Sonnet 5 |
| Max output | 128K tokens | — | — |
| Input price / 1M | $2 ↗ | — | — |
| Output price / 1M | $10 ↗ | — | — |
| Cached input / 1M | $0.2 ↗ | — | — |
| Input + output / 1M Lower is cheaper. List prices; batch, tool and regional fees excluded. | $12 | — | — |
| Modalities | text · vision | text · vision | Tie |
| Released | 2026-06-30 | 2025-07-09 | — |
| Cited benchmark scores | 36 | 17 | — |
Reliability
Provider status
Live from /statusMore matchups
Claude Sonnet 5 vs …
Models sharing the most benchmarksMore matchups
Grok 4 vs …
Models sharing the most benchmarksBuilt by Respan
Which one wins on your data?
Public benchmarks are a starting point. Run Claude Sonnet 5 and Grok 4 on your own prompts with Respan evals, or route to either through one gateway key with automatic failover.