# Claude Opus 4 vs Grok 4.5

> Grok 4.5 is the stronger model overall, scoring 55.0 to 43.1 on the Noometry Index.

- Canonical page: https://noometry.com/compare/claude-opus-4-vs-grok-4-5
- Last updated: 2026-10-10
- Shared benchmarks: 31

## Summary

- They share 31 benchmarks with published results for both. Claude Opus 4 scores higher in 1 category and Grok 4.5 in 9 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Grok 4.5 leads 56.1 to 27.3.
- The biggest single-benchmark swing is ARC-AGI-1: 35.7% for Claude Opus 4 and 87.2% for Grok 4.5.
- Grok 4.5 is cheaper at $2 / $6 per million input/output tokens, against $15 / $75 for Claude Opus 4.
- Grok 4.5 accepts more context: 500K tokens versus 200K.

## Snapshot

| | Claude Opus 4 | Grok 4.5 |
|---|---|---|
| Provider | Anthropic | xAI |
| Noometry Index | 43.1 | 55.0 |
| Rank | 100 | 25 |
| Context | 200K | 500K |
| Input $/M | $15 | $2 |
| Output $/M | $75 | $6 |
| Weights | Proprietary | Proprietary |

## Coding

- Claude Opus 4: 47.2 (#62)
- Grok 4.5: 52.2 (#35)

| Benchmark | Claude Opus 4 | Grok 4.5 |
|---|---|---|
| WeirdML | 43.7% | 46.4% |
| LMArena Coding | 1442 | 1474 |
| SWE-bench Verified | 70.7% | — |
| DeepSWE | — | 53.8% |
| FrontierCode | — | 42.4% |
| SWE-bench Verified (bash only) | 67.6% | — |
| Aider Polyglot | 72% | — |
| LMArena WebDev | — | 1553 |
| SciCode | — | 54.1% |
| GSO | 6.9% | — |
| ALE-Bench | — | 1,309 |
| AlgoTune | 1.33 | — |

## Agentic & Tool Use

- Claude Opus 4: 34.8 (#42)
- Grok 4.5: 44.4 (#17)

| Benchmark | Claude Opus 4 | Grok 4.5 |
|---|---|---|
| LMArena Search | 1127 | 1213 |
| APEX-Agents | — | 56.2% |
| τ²-bench Banking | — | 47.9% |
| Cybench | 38% | — |
| DeepResearch Bench | 46.8% | — |
| PostTrainBench | — | 23.4% |
| GBAEval | — | 65.4% |
| GDP.pdf | — | 14% |
| METR Time Horizons | 63.9% | — |
| Vending-Bench 2 | — | 3,887 |

## Reasoning

- Claude Opus 4: 27.3 (#121)
- Grok 4.5: 56.1 (#25)

| Benchmark | Claude Opus 4 | Grok 4.5 |
|---|---|---|
| ARC-AGI-2 | 8.6% | 52.6% |
| SimpleBench | 58.8% | 70% |
| Kagi LLM Benchmark | 74.3% | 83.5% |
| ARC-AGI-1 | 35.7% | 87.2% |
| CritPt | 0.3% | 15.4% |
| LMArena Hard Prompts | 1399 | 1462 |
| DTBench | 81.6% | 96.5% |
| LMCA | 37.4% | 45.2% |
| Epoch Capabilities Index | 142.67 | 153.92 |
| NYT Connections (extended) | — | 79.9% |
| Chess Puzzles | — | 36% |
| EnigmaEval | 5.6% | — |
| Surface Evolver Bench | — | 74.4% |
| ForecastBench | 61.1 | — |

## Math

- Claude Opus 4: 42.0 (#86)
- Grok 4.5: 60.9 (#35)

| Benchmark | Claude Opus 4 | Grok 4.5 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 64.4% | 97.8% |
| LMArena Math | 1390 | 1459 |
| FrontierMath (Tiers 1-3) | — | 57.2% |
| FrontierMath Tier 4 | — | 24.4% |
| ProofBench | — | 31% |
| Omni-MATH | 61.6% | — |
| MATH Level 5 | 85% | — |
| FrontierMath (Feb 2025 set) | 4.5% | — |
| FrontierMath Tier 4 (v1) | 4.2% | — |

## Knowledge

- Claude Opus 4: 44.0 (#88)
- Grok 4.5: 62.3 (#24)

| Benchmark | Claude Opus 4 | Grok 4.5 |
|---|---|---|
| GPQA Diamond | 76.3% | 93.4% |
| LMArena Expert | 1386 | 1466 |
| Humanity's Last Exam | 10.7% | — |
| SimpleQA Verified | — | 48.3% |
| MMLU-Pro | 87.5% | — |
| Confabulations | 15.9% | — |
| Vectara Hallucination Rate | 12% | — |
| GPQA (HELM) | 70.8% | — |

## Multimodal

- Claude Opus 4: 31.5 (#106)
- Grok 4.5: 37.6 (#72)

| Benchmark | Claude Opus 4 | Grok 4.5 |
|---|---|---|
| LMArena Vision | 1192 | 1288 |
| GeoBench | 49% | — |
| VPCT | 38% | — |
| Blueprint-Bench 2 | — | 27.3% |
| Furniture Assembly | — | 22.5% |
| LMArena Document | — | 1452 |

## Multilingual

- Claude Opus 4: 48.8 (#138)
- Grok 4.5: 54.4 (#42)

| Benchmark | Claude Opus 4 | Grok 4.5 |
|---|---|---|
| LMArena Non-English | 1362 | 1440 |
| LMArena Chinese | 1386 | 1496 |
| LMArena French | 1372 | 1456 |
| LMArena German | 1391 | 1446 |
| LMArena Japanese | 1331 | 1428 |
| LMArena Korean | 1321 | 1404 |
| LMArena Russian | 1392 | 1448 |
| LMArena Spanish | 1389 | 1450 |

## Instruction Following

- Claude Opus 4: 77.1 (#28)
- Grok 4.5: 76.0 (#48)

| Benchmark | Claude Opus 4 | Grok 4.5 |
|---|---|---|
| LMArena Instruction Following | 1406 | 1446 |
| IFEval | 91.8% | — |

## Long Context

- Claude Opus 4: 39.6 (#172)
- Grok 4.5: 44.8 (#56)

| Benchmark | Claude Opus 4 | Grok 4.5 |
|---|---|---|
| LMArena Longer Query | 1422 | 1463 |
| Fiction.LiveBench | 61.1% | — |

## Writing & Preference

- Claude Opus 4: 61.2 (#89)
- Grok 4.5: 65.8 (#42)

| Benchmark | Claude Opus 4 | Grok 4.5 |
|---|---|---|
| LMArena Text | 1377 | 1448 |
| LMArena Creative Writing | 1387 | 1442 |
| EQ-Bench Creative Writing | 1580 | 1579 |
| LMArena Multi-Turn | 1396 | 1456 |
| Short-Story Creative Writing | 83.6% | — |
| WildBench | 85.2% | — |

## FAQ

### Is Claude Opus 4 better than Grok 4.5?

Grok 4.5 is the stronger model overall, scoring 55.0 to 43.1 on the Noometry Index.

### Which is cheaper, Claude Opus 4 or Grok 4.5?

Grok 4.5 is cheaper. It lists at $2 per million input tokens and $6 per million output tokens; Claude Opus 4 lists at $15 and $75.

### Is Claude Opus 4 or Grok 4.5 better for coding?

Grok 4.5 scores higher on coding benchmarks: 52.2 versus 47.2 in the Noometry coding category.

### Which has the bigger context window?

Grok 4.5 does, with 500K tokens against 200K.

### How many benchmarks do Claude Opus 4 and Grok 4.5 share?

31 benchmarks have published results for both models. Claude Opus 4 has 56 scored results on Noometry and Grok 4.5 has 52.