# Claude Opus 5 vs Grok 4.6

> Claude Opus 5 is the stronger model overall, scoring 67.7 to 56.8 on the Noometry Index. Grok 4.6 costs 3.3× less per token, which makes it the better buy when Claude Opus 5's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/claude-opus-5-vs-grok-4-6
- Last updated: 2026-10-10
- Shared benchmarks: 49

## Summary

- They share 49 benchmarks with published results for both. Claude Opus 5 scores higher in 10 categories and Grok 4.6 in 0 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where Claude Opus 5 leads 86.1 to 66.8.
- The biggest single-benchmark swing is ProofBench: 99% for Claude Opus 5 and 51% for Grok 4.6.
- Grok 4.6 is cheaper at $2 / $6 per million input/output tokens, against $5 / $25 for Claude Opus 5.
- Claude Opus 5 accepts more context: 1M tokens versus 500K.

## Snapshot

| | Claude Opus 5 | Grok 4.6 |
|---|---|---|
| Provider | Anthropic | xAI |
| Noometry Index | 67.7 | 56.8 |
| Rank | 4 | 19 |
| Context | 1M | 500K |
| Input $/M | $5 | $2 |
| Output $/M | $25 | $6 |
| Weights | Proprietary | Proprietary |

## Coding

- Claude Opus 5: 67.0 (#5)
- Grok 4.6: 58.0 (#14)

| Benchmark | Claude Opus 5 | Grok 4.6 |
|---|---|---|
| DeepSWE | 73.6% | 67.5% |
| FrontierCode | 53.4% | 48% |
| CursorBench | 46.6% | 41.4% |
| LMArena WebDev | 1694 | 1620 |
| FrontierSWE | 52% | 25.3% |
| SciCode | 56.4% | 56.5% |
| WeirdML | 91.8% | 67.3% |
| LMArena Coding | 1533 | 1464 |
| ALE-Bench | 2,165 | 1,508 |

## Agentic & Tool Use

- Claude Opus 5: 55.5 (#1)
- Grok 4.6: 39.4 (#25)

| Benchmark | Claude Opus 5 | Grok 4.6 |
|---|---|---|
| APEX-Agents | 65.8% | 65.3% |
| GDP.pdf | 24% | 17.2% |
| Vending-Bench 2 | 11,182 | 9,047 |
| OSWorld 2.0 | 31.4% | — |
| τ²-bench Banking | 48.7% | — |
| PostTrainBench | 35% | — |
| BALROG | 63.4% | — |
| GBAEval | 79.6% | — |

## Reasoning

- Claude Opus 5: 77.1 (#3)
- Grok 4.6: 61.3 (#18)

| Benchmark | Claude Opus 5 | Grok 4.6 |
|---|---|---|
| ARC-AGI-2 | 90.4% | 67.1% |
| SimpleBench | 80.6% | 75.9% |
| NYT Connections (extended) | 94.3% | 80% |
| ARC-AGI-1 | 97.5% | 87.5% |
| CritPt | 29.1% | 19.7% |
| Chess Puzzles | 42% | 40% |
| EBR-Bench | 45.7% | 30.5% |
| LMArena Hard Prompts | 1525 | 1445 |
| Mystery Game Puzzles | 59% | 34% |
| DTBench | 97.9% | 97.3% |
| LMCA | 64.5% | 48.5% |
| Epoch Capabilities Index | 162.67 | 156.48 |
| Bench to the Future 3 | 0.12 | — |

## Math

- Claude Opus 5: 86.1 (#6)
- Grok 4.6: 66.8 (#19)

| Benchmark | Claude Opus 5 | Grok 4.6 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 85.6% | 66% |
| FrontierMath Tier 4 | 73.2% | 31.7% |
| OTIS Mock AIME 2024-2025 | 98.9% | 99.2% |
| ProofBench | 99% | 51% |
| LMArena Math | 1533 | 1416 |

## Knowledge

- Claude Opus 5: 66.8 (#8)
- Grok 4.6: 63.3 (#18)

| Benchmark | Claude Opus 5 | Grok 4.6 |
|---|---|---|
| GPQA Diamond | 93.9% | 94% |
| SimpleQA Verified | 59.9% | 49.3% |
| LMArena Expert | 1558 | 1463 |

## Multimodal

- Claude Opus 5: 51.1 (#6)
- Grok 4.6: 43.9 (#18)

| Benchmark | Claude Opus 5 | Grok 4.6 |
|---|---|---|
| LMArena Vision | 1321 | 1265 |
| Blueprint-Bench 2 | 30.4% | 33.2% |
| Furniture Assembly | 60.8% | 40% |
| LMArena Document | 1516 | 1452 |

## Multilingual

- Claude Opus 5: 58.8 (#3)
- Grok 4.6: 53.3 (#67)

| Benchmark | Claude Opus 5 | Grok 4.6 |
|---|---|---|
| LMArena Non-English | 1500 | 1422 |
| LMArena Chinese | 1575 | 1479 |
| LMArena French | 1522 | 1458 |
| LMArena German | 1522 | 1425 |
| LMArena Japanese | 1510 | 1383 |
| LMArena Korean | 1526 | 1395 |
| LMArena Russian | 1507 | 1427 |
| LMArena Spanish | 1519 | 1408 |

## Instruction Following

- Claude Opus 5: 79.3 (#5)
- Grok 4.6: 75.6 (#59)

| Benchmark | Claude Opus 5 | Grok 4.6 |
|---|---|---|
| LMArena Instruction Following | 1516 | 1431 |

## Long Context

- Claude Opus 5: 46.5 (#20)
- Grok 4.6: 44.4 (#62)

| Benchmark | Claude Opus 5 | Grok 4.6 |
|---|---|---|
| LMArena Longer Query | 1516 | 1453 |

## Writing & Preference

- Claude Opus 5: 79.3 (#2)
- Grok 4.6: 62.5 (#77)

| Benchmark | Claude Opus 5 | Grok 4.6 |
|---|---|---|
| LMArena Text | 1506 | 1427 |
| LMArena Creative Writing | 1491 | 1430 |
| LMArena Multi-Turn | 1496 | 1422 |
| EQ-Bench Creative Writing | 2133 | — |
| EQ-Bench 4 | 1385 | — |

## FAQ

### Is Claude Opus 5 better than Grok 4.6?

Claude Opus 5 is the stronger model overall, scoring 67.7 to 56.8 on the Noometry Index. Grok 4.6 costs 3.3× less per token, which makes it the better buy when Claude Opus 5's lead doesn't matter for your workload.

### Which is cheaper, Claude Opus 5 or Grok 4.6?

Grok 4.6 is cheaper. It lists at $2 per million input tokens and $6 per million output tokens; Claude Opus 5 lists at $5 and $25.

### Is Claude Opus 5 or Grok 4.6 better for coding?

Claude Opus 5 scores higher on coding benchmarks: 67.0 versus 58.0 in the Noometry coding category.

### Which has the bigger context window?

Claude Opus 5 does, with 1M tokens against 500K.

### How many benchmarks do Claude Opus 5 and Grok 4.6 share?

49 benchmarks have published results for both models. Claude Opus 5 has 57 scored results on Noometry and Grok 4.6 has 49.