# GPT-4 vs Grok-2 (Dec 2024)

> Grok-2 (Dec 2024) is the stronger model overall, scoring 33.7 to 29.1 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gpt-4-vs-grok-2
- Last updated: 2026-10-11
- Shared benchmarks: 23

## Summary

- They share 23 benchmarks with published results for both. GPT-4 scores higher in 1 category and Grok-2 (Dec 2024) in 7 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Grok-2 (Dec 2024) leads 48.6 to 34.9.
- The biggest single-benchmark swing is MATH Level 5: 23% for GPT-4 and 63.5% for Grok-2 (Dec 2024).

## Snapshot

| | GPT-4 | Grok-2 (Dec 2024) |
|---|---|---|
| Provider | OpenAI | xAI |
| Noometry Index | 29.1 | 33.7 |
| Rank | 316 | 239 |
| Context | 8K | — |
| Input $/M | $30 | — |
| Output $/M | $60 | — |
| Weights | Proprietary | Proprietary |

## Coding

- GPT-4: 31.6 (#283)
- Grok-2 (Dec 2024): 33.3 (#258)

| Benchmark | GPT-4 | Grok-2 (Dec 2024) |
|---|---|---|
| WeirdML | 12.4% | 22.2% |
| LMArena Coding | 1254 | 1287 |
| BigCodeBench Instruct | 46% | — |
| LiveBench Coding | — | 46.4% |
| BigCodeBench Complete | 57.2% | — |
| HumanEval+ | 79.3% | — |

## Agentic & Tool Use

- GPT-4: —
- Grok-2 (Dec 2024): —

| Benchmark | GPT-4 | Grok-2 (Dec 2024) |
|---|---|---|
| METR Time Horizons | 36.1% | — |

## Reasoning

- GPT-4: 17.8 (#289)
- Grok-2 (Dec 2024): 16.9 (#299)

| Benchmark | GPT-4 | Grok-2 (Dec 2024) |
|---|---|---|
| LMArena Hard Prompts | 1241 | 1272 |
| DTBench | 62.7% | 65.2% |
| Epoch Capabilities Index | 125.89 | 130.48 |
| SimpleBench | — | 22.7% |
| Chess Puzzles | 4% | — |
| LiveBench Reasoning | — | 54.8% |
| Mystery Game Puzzles | 12% | — |
| LiveBench Data Analysis | — | 54.5% |
| LMCA | 17.1% | — |
| BIG-Bench Hard | 75.1% | — |
| ForecastBench | 57.8 | — |
| HellaSwag | 95.3% | — |
| LiveBench | — | 54.3% |
| WinoGrande | 87.5% | — |

## Math

- GPT-4: 10.8 (#309)
- Grok-2 (Dec 2024): 20.8 (#284)

| Benchmark | GPT-4 | Grok-2 (Dec 2024) |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.1% | 11.5% |
| LMArena Math | 1269 | 1283 |
| MATH Level 5 | 23% | 63.5% |
| LiveBench Math | — | 54.9% |
| FrontierMath (Feb 2025 set) | — | 0.7% |
| GSM8K | 92% | — |

## Knowledge

- GPT-4: 18.4 (#282)
- Grok-2 (Dec 2024): 29.8 (#233)

| Benchmark | GPT-4 | Grok-2 (Dec 2024) |
|---|---|---|
| GPQA Diamond | 35.7% | 53.8% |
| LMArena Expert | 1211 | 1254 |
| Confabulations | — | 20.1% |
| MMLU | 86.4% | — |
| TriviaQA | 84.8% | — |

## Multilingual

- GPT-4: 40.6 (#215)
- Grok-2 (Dec 2024): 43.1 (#188)

| Benchmark | GPT-4 | Grok-2 (Dec 2024) |
|---|---|---|
| LMArena Non-English | 1246 | 1282 |
| LMArena Chinese | 1242 | 1289 |
| LMArena French | 1283 | 1318 |
| LMArena German | 1251 | 1287 |
| LMArena Japanese | 1209 | 1244 |
| LMArena Korean | 1184 | 1237 |
| LMArena Russian | 1251 | 1286 |
| LMArena Spanish | 1261 | 1281 |

## Instruction Following

- GPT-4: 65.3 (#222)
- Grok-2 (Dec 2024): 66.9 (#202)

| Benchmark | GPT-4 | Grok-2 (Dec 2024) |
|---|---|---|
| LMArena Instruction Following | 1241 | 1270 |
| LiveBench Instruction Following | — | 69.6% |

## Long Context

- GPT-4: 37.7 (#212)
- Grok-2 (Dec 2024): 38.8 (#190)

| Benchmark | GPT-4 | Grok-2 (Dec 2024) |
|---|---|---|
| LMArena Longer Query | 1244 | 1276 |

## Writing & Preference

- GPT-4: 34.9 (#268)
- Grok-2 (Dec 2024): 48.6 (#198)

| Benchmark | GPT-4 | Grok-2 (Dec 2024) |
|---|---|---|
| LMArena Text | 1263 | 1305 |
| LMArena Creative Writing | 1244 | 1284 |
| LMArena Multi-Turn | 1257 | 1290 |
| Short-Story Creative Writing | — | 63.6% |
| EQ-Bench Creative Writing | 752 | — |
| LiveBench Language | — | 45.6% |

## FAQ

### Is GPT-4 better than Grok-2 (Dec 2024)?

Grok-2 (Dec 2024) is the stronger model overall, scoring 33.7 to 29.1 on the Noometry Index.

### Is GPT-4 or Grok-2 (Dec 2024) better for coding?

Grok-2 (Dec 2024) scores higher on coding benchmarks: 33.3 versus 31.6 in the Noometry coding category.

### How many benchmarks do GPT-4 and Grok-2 (Dec 2024) share?

23 benchmarks have published results for both models. GPT-4 has 38 scored results on Noometry and Grok-2 (Dec 2024) has 34.