# Claude Sonnet 4.5 vs Grok 4.20 (Non-Reasoning)

> Grok 4.20 (Non-Reasoning) is the stronger model overall, scoring 48.6 to 44.1 on the Noometry Index.

- Canonical page: https://noometry.com/compare/claude-sonnet-4-5-vs-grok-4-20
- Last updated: 2026-10-12
- Shared benchmarks: 41

## Summary

- They share 41 benchmarks with published results for both. Claude Sonnet 4.5 scores higher in 5 categories and Grok 4.20 (Non-Reasoning) in 5 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Grok 4.20 (Non-Reasoning) leads 52.3 to 26.9.
- The biggest single-benchmark swing is ARC-AGI-2: 13.6% for Claude Sonnet 4.5 and 65.1% for Grok 4.20 (Non-Reasoning).
- Grok 4.20 (Non-Reasoning) is cheaper at $1.25 / $2.50 per million input/output tokens, against $3 / $15 for Claude Sonnet 4.5.
- Grok 4.20 (Non-Reasoning) accepts more context: 1M tokens versus 200K.

## Snapshot

| | Claude Sonnet 4.5 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| Provider | Anthropic | xAI |
| Noometry Index | 44.1 | 48.6 |
| Rank | 81 | 54 |
| Context | 200K | 1M |
| Input $/M | $3 | $1.25 |
| Output $/M | $15 | $2.50 |
| Weights | Proprietary | Proprietary |

## Coding

- Claude Sonnet 4.5: 47.3 (#61)
- Grok 4.20 (Non-Reasoning): 42.1 (#112)

| Benchmark | Claude Sonnet 4.5 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena WebDev | 1393 | 1375 |
| WeirdML | 47.7% | 52.3% |
| LMArena Coding | 1489 | 1459 |
| ALE-Bench | 796.15 | 1,150 |
| SWE-bench Verified | 71.3% | — |
| SWE-bench Verified (bash only) | 71.4% | — |
| SWE-bench Multilingual | 67% | — |
| SciCode | 44.7% | — |
| GSO | 14.7% | — |
| AlgoTune | 1.52 | — |

## Agentic & Tool Use

- Claude Sonnet 4.5: 38.3 (#32)
- Grok 4.20 (Non-Reasoning): 34.4 (#46)

| Benchmark | Claude Sonnet 4.5 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| Terminal-Bench | 46.5% | 57.3% |
| τ²-bench Banking | 25.3% | 18% |
| LMArena Search | 1159 | 1189 |
| Vending-Bench 2 | 3,839 | 4,663 |
| Berkeley Function Calling Leaderboard | 73.2% | — |
| GDPval | 42.5% | — |
| Remote Labor Index | 2.1% | — |
| τ²-bench Airline | 72% | — |
| τ²-bench Retail | 72.4% | — |
| τ²-bench Telecom | 84.9% | — |
| Cybench | 60% | — |
| DeepResearch Bench | 52.6% | — |
| OSWorld | 62.9% | — |
| METR Time Horizons | 67.4% | — |

## Reasoning

- Claude Sonnet 4.5: 26.9 (#125)
- Grok 4.20 (Non-Reasoning): 52.3 (#32)

| Benchmark | Claude Sonnet 4.5 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| ARC-AGI-2 | 13.6% | 65.1% |
| Kagi LLM Benchmark | 57.9% | 75% |
| NYT Connections (extended) | 37.3% | 85.4% |
| ARC-AGI-1 | 63.7% | 89.5% |
| Chess Puzzles | 12% | 24% |
| LMArena Hard Prompts | 1462 | 1451 |
| DTBench | 83.2% | 90.1% |
| LMCA | 38.8% | 38.7% |
| Epoch Capabilities Index | 146.84 | 151.98 |
| ForecastBench | 61.9 | 61.4 |
| SimpleBench | 54.3% | — |
| CritPt | 1.1% | — |
| EnigmaEval | 6% | — |
| Thematic Generalization | — | 63.8% |
| EBR-Bench | 2.4% | — |
| Mystery Game Puzzles | 17% | — |

## Math

- Claude Sonnet 4.5: 32.3 (#216)
- Grok 4.20 (Non-Reasoning): 48.2 (#65)

| Benchmark | Claude Sonnet 4.5 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| FrontierMath (Tiers 1-3) | 23.9% | 44.9% |
| FrontierMath Tier 4 | 2.4% | 17.1% |
| OTIS Mock AIME 2024-2025 | 77.8% | 92.2% |
| ProofBench | 19% | 14% |
| LMArena Math | 1449 | 1455 |
| Omni-MATH | 55.3% | — |
| MATH Level 5 | 97.7% | — |
| FrontierMath (Feb 2025 set) | 15.2% | — |
| FrontierMath Tier 4 (v1) | 4.2% | — |

## Knowledge

- Claude Sonnet 4.5: 48.4 (#76)
- Grok 4.20 (Non-Reasoning): 52.8 (#60)

| Benchmark | Claude Sonnet 4.5 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| GPQA Diamond | 82.3% | 89.3% |
| SimpleQA Verified | 30.7% | 30.2% |
| LMArena Expert | 1482 | 1439 |
| Humanity's Last Exam | 13.7% | — |
| MMLU-Pro | 86.9% | — |
| Vectara Hallucination Rate | 12% | — |
| GPQA (HELM) | 68.6% | — |

## Multimodal

- Claude Sonnet 4.5: 34.8 (#89)
- Grok 4.20 (Non-Reasoning): 33.3 (#98)

| Benchmark | Claude Sonnet 4.5 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena Document | 1450 | 1416 |
| LMArena Vision | — | 1263 |
| VPCT | 39.8% | — |
| Blueprint-Bench 2 | — | 0% |

## Multilingual

- Claude Sonnet 4.5: 53.4 (#69)
- Grok 4.20 (Non-Reasoning): 54.5 (#40)

| Benchmark | Claude Sonnet 4.5 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena Non-English | 1425 | 1441 |
| LMArena Chinese | 1459 | 1481 |
| LMArena French | 1458 | 1476 |
| LMArena German | 1427 | 1465 |
| LMArena Japanese | 1390 | 1449 |
| LMArena Korean | 1403 | 1417 |
| LMArena Russian | 1437 | 1458 |
| LMArena Spanish | 1457 | 1443 |

## Instruction Following

- Claude Sonnet 4.5: 75.0 (#78)
- Grok 4.20 (Non-Reasoning): 74.8 (#83)

| Benchmark | Claude Sonnet 4.5 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena Instruction Following | 1459 | 1420 |
| IFEval | 85% | — |

## Long Context

- Claude Sonnet 4.5: 45.2 (#46)
- Grok 4.20 (Non-Reasoning): 45.5 (#34)

| Benchmark | Claude Sonnet 4.5 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena Longer Query | 1476 | 1437 |
| CL-bench | — | 22.2% |
| CL-bench Life | — | 11.9% |

## Writing & Preference

- Claude Sonnet 4.5: 66.5 (#34)
- Grok 4.20 (Non-Reasoning): 65.7 (#44)

| Benchmark | Claude Sonnet 4.5 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena Text | 1439 | 1451 |
| LMArena Creative Writing | 1442 | 1438 |
| EQ-Bench Creative Writing | 1678 | 1574 |
| LMArena Multi-Turn | 1465 | 1456 |
| WildBench | 85.4% | — |

## FAQ

### Is Claude Sonnet 4.5 better than Grok 4.20 (Non-Reasoning)?

Grok 4.20 (Non-Reasoning) is the stronger model overall, scoring 48.6 to 44.1 on the Noometry Index.

### Which is cheaper, Claude Sonnet 4.5 or Grok 4.20 (Non-Reasoning)?

Grok 4.20 (Non-Reasoning) is cheaper. It lists at $1.25 per million input tokens and $2.50 per million output tokens; Claude Sonnet 4.5 lists at $3 and $15.

### Is Claude Sonnet 4.5 or Grok 4.20 (Non-Reasoning) better for coding?

Claude Sonnet 4.5 scores higher on coding benchmarks: 47.3 versus 42.1 in the Noometry coding category.

### Which has the bigger context window?

Grok 4.20 (Non-Reasoning) does, with 1M tokens against 200K.

### How many benchmarks do Claude Sonnet 4.5 and Grok 4.20 (Non-Reasoning) share?

41 benchmarks have published results for both models. Claude Sonnet 4.5 has 73 scored results on Noometry and Grok 4.20 (Non-Reasoning) has 46.