# GPT-4 vs Grok 4

> Grok 4 is the stronger model overall, scoring 48.1 to 29.1 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gpt-4-vs-grok-4
- Last updated: 2026-10-11
- Shared benchmarks: 24

## Summary

- They share 24 benchmarks with published results for both. GPT-4 scores higher in 0 categories and Grok 4 in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Grok 4 leads 48.4 to 10.8.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 1.1% for GPT-4 and 84% for Grok 4.

## Snapshot

| | GPT-4 | Grok 4 |
|---|---|---|
| Provider | OpenAI | xAI |
| Noometry Index | 29.1 | 48.1 |
| Rank | 316 | 56 |
| Context | 8K | — |
| Input $/M | $30 | — |
| Output $/M | $60 | — |
| Weights | Proprietary | Proprietary |

## Coding

- GPT-4: 31.6 (#283)
- Grok 4: 50.3 (#46)

| Benchmark | GPT-4 | Grok 4 |
|---|---|---|
| WeirdML | 12.4% | 45.7% |
| LMArena Coding | 1254 | 1408 |
| Aider Polyglot | — | 79.6% |
| BigCodeBench Instruct | 46% | — |
| BigCodeBench Complete | 57.2% | — |
| HumanEval+ | 79.3% | — |

## Agentic & Tool Use

- GPT-4: —
- Grok 4: 32.3 (#68)

| Benchmark | GPT-4 | Grok 4 |
|---|---|---|
| METR Time Horizons | 36.1% | 66.6% |
| Terminal-Bench | — | 27.2% |
| Berkeley Function Calling Leaderboard | — | 63% |
| GDPval | — | 21.1% |
| Cybench | — | 43% |
| DeepResearch Bench | — | 47.3% |
| BALROG | — | 43.6% |
| LMArena Search | — | 1142 |

## Reasoning

- GPT-4: 17.8 (#289)
- Grok 4: 36.7 (#65)

| Benchmark | GPT-4 | Grok 4 |
|---|---|---|
| Chess Puzzles | 4% | 28% |
| LMArena Hard Prompts | 1241 | 1409 |
| Epoch Capabilities Index | 125.89 | 146.44 |
| ForecastBench | 57.8 | 60.9 |
| ARC-AGI-2 | — | 16% |
| SimpleBench | — | 60.5% |
| Kagi LLM Benchmark | — | 73.6% |
| ARC-AGI-1 | — | 66.7% |
| Mystery Game Puzzles | 12% | — |
| DTBench | 62.7% | — |
| LMCA | 17.1% | — |
| BIG-Bench Hard | 75.1% | — |
| HellaSwag | 95.3% | — |
| WinoGrande | 87.5% | — |

## Math

- GPT-4: 10.8 (#309)
- Grok 4: 48.4 (#64)

| Benchmark | GPT-4 | Grok 4 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.1% | 84% |
| LMArena Math | 1269 | 1422 |
| Omni-MATH | — | 60.3% |
| MATH Level 5 | 23% | — |
| FrontierMath (Feb 2025 set) | — | 19.7% |
| FrontierMath Tier 4 (v1) | — | 2.1% |
| GSM8K | 92% | — |

## Knowledge

- GPT-4: 18.4 (#282)
- Grok 4: 53.8 (#55)

| Benchmark | GPT-4 | Grok 4 |
|---|---|---|
| GPQA Diamond | 35.7% | 87% |
| LMArena Expert | 1211 | 1415 |
| MMLU-Pro | — | 85.1% |
| Confabulations | — | 12.4% |
| GPQA (HELM) | — | 72.7% |
| MMLU | 86.4% | — |
| TriviaQA | 84.8% | — |

## Multimodal

- GPT-4: —
- Grok 4: 33.7 (#94)

| Benchmark | GPT-4 | Grok 4 |
|---|---|---|
| LMArena Vision | — | 1210 |
| GeoBench | — | 45% |

## Multilingual

- GPT-4: 40.6 (#215)
- Grok 4: 51.8 (#103)

| Benchmark | GPT-4 | Grok 4 |
|---|---|---|
| LMArena Non-English | 1246 | 1403 |
| LMArena Chinese | 1242 | 1427 |
| LMArena French | 1283 | 1418 |
| LMArena German | 1251 | 1429 |
| LMArena Japanese | 1209 | 1394 |
| LMArena Korean | 1184 | 1377 |
| LMArena Russian | 1251 | 1410 |
| LMArena Spanish | 1261 | 1420 |

## Instruction Following

- GPT-4: 65.3 (#222)
- Grok 4: 79.2 (#5)

| Benchmark | GPT-4 | Grok 4 |
|---|---|---|
| LMArena Instruction Following | 1241 | 1387 |
| IFEval | — | 94.9% |

## Long Context

- GPT-4: 37.7 (#212)
- Grok 4: 63.1 (#4)

| Benchmark | GPT-4 | Grok 4 |
|---|---|---|
| LMArena Longer Query | 1244 | 1409 |
| Fiction.LiveBench | — | 94.4% |

## Writing & Preference

- GPT-4: 34.9 (#268)
- Grok 4: 58.5 (#116)

| Benchmark | GPT-4 | Grok 4 |
|---|---|---|
| LMArena Text | 1263 | 1411 |
| LMArena Creative Writing | 1244 | 1397 |
| LMArena Multi-Turn | 1257 | 1416 |
| Short-Story Creative Writing | — | 76.9% |
| EQ-Bench Creative Writing | 752 | — |
| WildBench | — | 79.7% |

## FAQ

### Is GPT-4 better than Grok 4?

Grok 4 is the stronger model overall, scoring 48.1 to 29.1 on the Noometry Index.

### Is GPT-4 or Grok 4 better for coding?

Grok 4 scores higher on coding benchmarks: 50.3 versus 31.6 in the Noometry coding category.

### How many benchmarks do GPT-4 and Grok 4 share?

24 benchmarks have published results for both models. GPT-4 has 38 scored results on Noometry and Grok 4 has 48.