# Claude Opus 4.6 vs Grok 4.20 (Non-Reasoning)

> Claude Opus 4.6 is the stronger model overall, scoring 58.2 to 48.6 on the Noometry Index. Grok 4.20 (Non-Reasoning) costs 6.4× less per token, which makes it the better buy when Claude Opus 4.6's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/claude-opus-4-6-vs-grok-4-20
- Last updated: 2026-10-12
- Shared benchmarks: 45

## Summary

- They share 45 benchmarks with published results for both. Claude Opus 4.6 scores higher in 10 categories and Grok 4.20 (Non-Reasoning) in 0 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in agentic & tool use, where Claude Opus 4.6 leads 51.1 to 34.4.
- The biggest single-benchmark swing is ProofBench: 50% for Claude Opus 4.6 and 14% for Grok 4.20 (Non-Reasoning).
- Grok 4.20 (Non-Reasoning) is cheaper at $1.25 / $2.50 per million input/output tokens, against $5 / $25 for Claude Opus 4.6.

## Snapshot

| | Claude Opus 4.6 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| Provider | Anthropic | xAI |
| Noometry Index | 58.2 | 48.6 |
| Rank | 20 | 54 |
| Context | 1M | 1M |
| Input $/M | $5 | $1.25 |
| Output $/M | $25 | $2.50 |
| Weights | Proprietary | Proprietary |

## Coding

- Claude Opus 4.6: 57.2 (#20)
- Grok 4.20 (Non-Reasoning): 42.1 (#112)

| Benchmark | Claude Opus 4.6 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena WebDev | 1547 | 1375 |
| WeirdML | 78% | 52.3% |
| LMArena Coding | 1536 | 1459 |
| ALE-Bench | 996.5 | 1,150 |
| SWE-bench Verified | 78.7% | — |
| FrontierCode | 26.6% | — |
| SWE-bench Verified (bash only) | 75.6% | — |
| SWE-bench Multilingual | 72% | — |
| GSO | 41.2% | — |
| AlgoTune | 1.47 | — |

## Agentic & Tool Use

- Claude Opus 4.6: 51.1 (#4)
- Grok 4.20 (Non-Reasoning): 34.4 (#46)

| Benchmark | Claude Opus 4.6 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| Terminal-Bench | 79.8% | 57.3% |
| τ²-bench Banking | 27.3% | 18% |
| LMArena Search | 1253 | 1189 |
| Vending-Bench 2 | 8,018 | 4,663 |
| APEX-Agents | 46.3% | — |
| Remote Labor Index | 4.2% | — |
| Cybench | 93% | — |
| DeepResearch Bench | 55.3% | — |
| GBAEval | 44.1% | — |
| METR Time Horizons | 78.9% | — |

## Reasoning

- Claude Opus 4.6: 57.8 (#23)
- Grok 4.20 (Non-Reasoning): 52.3 (#32)

| Benchmark | Claude Opus 4.6 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| ARC-AGI-2 | 69.2% | 65.1% |
| Kagi LLM Benchmark | 83.6% | 75% |
| NYT Connections (extended) | 92.1% | 85.4% |
| ARC-AGI-1 | 94% | 89.5% |
| Chess Puzzles | 17% | 24% |
| Thematic Generalization | 80.6% | 63.8% |
| LMArena Hard Prompts | 1527 | 1451 |
| DTBench | 91.2% | 90.1% |
| LMCA | 55.8% | 38.7% |
| Epoch Capabilities Index | 155.24 | 151.98 |
| ForecastBench | 60 | 61.4 |
| SimpleBench | 67.6% | — |
| EnigmaEval | 7.6% | — |
| EBR-Bench | 12.7% | — |
| Mystery Game Puzzles | 25% | — |

## Math

- Claude Opus 4.6: 63.0 (#31)
- Grok 4.20 (Non-Reasoning): 48.2 (#65)

| Benchmark | Claude Opus 4.6 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| FrontierMath (Tiers 1-3) | 66% | 44.9% |
| FrontierMath Tier 4 | 26.8% | 17.1% |
| OTIS Mock AIME 2024-2025 | 94.4% | 92.2% |
| ProofBench | 50% | 14% |
| LMArena Math | 1519 | 1455 |
| MathArena Final-Answer Competitions | 78.5% | — |
| FrontierMath (Feb 2025 set) | 40.7% | — |
| FrontierMath Tier 4 (v1) | 22.9% | — |

## Knowledge

- Claude Opus 4.6: 61.9 (#26)
- Grok 4.20 (Non-Reasoning): 52.8 (#60)

| Benchmark | Claude Opus 4.6 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| GPQA Diamond | 90.5% | 89.3% |
| SimpleQA Verified | 47% | 30.2% |
| LMArena Expert | 1546 | 1439 |
| Humanity's Last Exam | 34.4% | — |
| Vectara Hallucination Rate | 12.2% | — |

## Multimodal

- Claude Opus 4.6: 37.3 (#74)
- Grok 4.20 (Non-Reasoning): 33.3 (#98)

| Benchmark | Claude Opus 4.6 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena Vision | 1316 | 1263 |
| LMArena Document | 1507 | 1416 |
| Blueprint-Bench 2 | — | 0% |
| Furniture Assembly | 28.3% | — |

## Multilingual

- Claude Opus 4.6: 57.9 (#6)
- Grok 4.20 (Non-Reasoning): 54.5 (#40)

| Benchmark | Claude Opus 4.6 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena Non-English | 1489 | 1441 |
| LMArena Chinese | 1551 | 1481 |
| LMArena French | 1513 | 1476 |
| LMArena German | 1502 | 1465 |
| LMArena Japanese | 1484 | 1449 |
| LMArena Korean | 1464 | 1417 |
| LMArena Russian | 1497 | 1458 |
| LMArena Spanish | 1510 | 1443 |

## Instruction Following

- Claude Opus 4.6: 79.5 (#4)
- Grok 4.20 (Non-Reasoning): 74.8 (#83)

| Benchmark | Claude Opus 4.6 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena Instruction Following | 1523 | 1420 |

## Long Context

- Claude Opus 4.6: 48.1 (#13)
- Grok 4.20 (Non-Reasoning): 45.5 (#34)

| Benchmark | Claude Opus 4.6 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| CL-bench | 20.7% | 22.2% |
| CL-bench Life | 17% | 11.9% |
| LMArena Longer Query | 1520 | 1437 |

## Writing & Preference

- Claude Opus 4.6: 73.5 (#10)
- Grok 4.20 (Non-Reasoning): 65.7 (#44)

| Benchmark | Claude Opus 4.6 | Grok 4.20 (Non-Reasoning) |
|---|---|---|
| LMArena Text | 1503 | 1451 |
| LMArena Creative Writing | 1505 | 1438 |
| EQ-Bench Creative Writing | 1809 | 1574 |
| LMArena Multi-Turn | 1513 | 1456 |
| EQ-Bench 4 | 1223 | — |

## FAQ

### Is Claude Opus 4.6 better than Grok 4.20 (Non-Reasoning)?

Claude Opus 4.6 is the stronger model overall, scoring 58.2 to 48.6 on the Noometry Index. Grok 4.20 (Non-Reasoning) costs 6.4× less per token, which makes it the better buy when Claude Opus 4.6's lead doesn't matter for your workload.

### Which is cheaper, Claude Opus 4.6 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 Opus 4.6 lists at $5 and $25.

### Is Claude Opus 4.6 or Grok 4.20 (Non-Reasoning) better for coding?

Claude Opus 4.6 scores higher on coding benchmarks: 57.2 versus 42.1 in the Noometry coding category.

### Which has the bigger context window?

Both accept 1M tokens.

### How many benchmarks do Claude Opus 4.6 and Grok 4.20 (Non-Reasoning) share?

45 benchmarks have published results for both models. Claude Opus 4.6 has 68 scored results on Noometry and Grok 4.20 (Non-Reasoning) has 46.