# GPT-4 Turbo vs Grok 3

> Grok 3 is the stronger model overall, scoring 39.6 to 30.3 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gpt-4-turbo-vs-grok-3
- Last updated: 2026-10-10
- Shared benchmarks: 24

## Summary

- They share 24 benchmarks with published results for both. GPT-4 Turbo scores higher in 1 category and Grok 3 in 7 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where Grok 3 leads 36.2 to 8.6.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 6.7% for GPT-4 Turbo and 55.6% for Grok 3.

## Snapshot

| | GPT-4 Turbo | Grok 3 |
|---|---|---|
| Provider | OpenAI | xAI |
| Noometry Index | 30.3 | 39.6 |
| Rank | 281 | 150 |
| Context | 128K | — |
| Input $/M | $10 | — |
| Output $/M | $30 | — |
| Weights | Proprietary | Proprietary |

## Coding

- GPT-4 Turbo: 33.4 (#247)
- Grok 3: 41.5 (#113)

| Benchmark | GPT-4 Turbo | Grok 3 |
|---|---|---|
| WeirdML | 18% | 37.2% |
| LMArena Coding | 1268 | 1432 |
| Aider Polyglot | — | 53.3% |
| BigCodeBench Instruct | 48.2% | — |
| BigCodeBench Complete | 58.2% | — |
| HumanEval+ | 86.6% | — |
| MBPP+ | 73.3% | — |

## Agentic & Tool Use

- GPT-4 Turbo: —
- Grok 3: 30.2 (#69)

| Benchmark | GPT-4 Turbo | Grok 3 |
|---|---|---|
| BALROG | — | 29.5% |
| METR Time Horizons | 36.7% | — |

## Reasoning

- GPT-4 Turbo: 14.9 (#311)
- Grok 3: 13.8 (#322)

| Benchmark | GPT-4 Turbo | Grok 3 |
|---|---|---|
| SimpleBench | 25.1% | 36.1% |
| LMArena Hard Prompts | 1251 | 1433 |
| Epoch Capabilities Index | 127.25 | 138.34 |
| ARC-AGI-2 | — | 0% |
| Kagi LLM Benchmark | — | 61.3% |
| ARC-AGI-1 | — | 5.5% |
| Chess Puzzles | 6% | — |
| DTBench | 61.6% | — |
| LMCA | 9.8% | — |
| ForecastBench | 59.4 | — |

## Math

- GPT-4 Turbo: 8.6 (#316)
- Grok 3: 36.2 (#122)

| Benchmark | GPT-4 Turbo | Grok 3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 6.7% | 55.6% |
| LMArena Math | 1272 | 1390 |
| MATH Level 5 | 46.7% | 88.7% |
| FrontierMath (Tiers 1-3) | 0.7% | — |
| Omni-MATH | — | 46.4% |
| FrontierMath (Feb 2025 set) | — | 3.8% |
| FrontierMath Tier 4 (v1) | — | 0% |

## Knowledge

- GPT-4 Turbo: 24.3 (#263)
- Grok 3: 46.0 (#76)

| Benchmark | GPT-4 Turbo | Grok 3 |
|---|---|---|
| GPQA Diamond | 46.6% | 75.8% |
| Confabulations | 28.4% | 14.2% |
| LMArena Expert | 1223 | 1420 |
| MMLU-Pro | — | 78.8% |
| Vectara Hallucination Rate | — | 5.8% |
| GPQA (HELM) | — | 65% |
| MMLU | 81.3% | — |

## Multimodal

- GPT-4 Turbo: 30.0 (#104)
- Grok 3: —

| Benchmark | GPT-4 Turbo | Grok 3 |
|---|---|---|
| LMArena Vision | 1090 | — |

## Multilingual

- GPT-4 Turbo: 40.7 (#211)
- Grok 3: 52.5 (#81)

| Benchmark | GPT-4 Turbo | Grok 3 |
|---|---|---|
| LMArena Non-English | 1245 | 1411 |
| LMArena Chinese | 1242 | 1448 |
| LMArena French | 1275 | 1460 |
| LMArena German | 1259 | 1431 |
| LMArena Japanese | 1194 | 1389 |
| LMArena Korean | 1187 | 1374 |
| LMArena Russian | 1259 | 1416 |
| LMArena Spanish | 1260 | 1417 |

## Instruction Following

- GPT-4 Turbo: 66.0 (#211)
- Grok 3: 75.2 (#69)

| Benchmark | GPT-4 Turbo | Grok 3 |
|---|---|---|
| LMArena Instruction Following | 1250 | 1409 |
| IFEval | — | 88.4% |

## Long Context

- GPT-4 Turbo: 38.0 (#200)
- Grok 3: 38.7 (#187)

| Benchmark | GPT-4 Turbo | Grok 3 |
|---|---|---|
| LMArena Longer Query | 1254 | 1439 |
| Fiction.LiveBench | — | 58.3% |

## Writing & Preference

- GPT-4 Turbo: 47.9 (#201)
- Grok 3: 56.0 (#136)

| Benchmark | GPT-4 Turbo | Grok 3 |
|---|---|---|
| LMArena Text | 1272 | 1426 |
| LMArena Creative Writing | 1269 | 1414 |
| LMArena Multi-Turn | 1267 | 1424 |
| Short-Story Creative Writing | — | 76.4% |
| EQ-Bench Creative Writing | — | 1186 |
| WildBench | — | 84.9% |

## FAQ

### Is GPT-4 Turbo better than Grok 3?

Grok 3 is the stronger model overall, scoring 39.6 to 30.3 on the Noometry Index.

### Is GPT-4 Turbo or Grok 3 better for coding?

Grok 3 scores higher on coding benchmarks: 41.5 versus 33.4 in the Noometry coding category.

### How many benchmarks do GPT-4 Turbo and Grok 3 share?

24 benchmarks have published results for both models. GPT-4 Turbo has 36 scored results on Noometry and Grok 3 has 40.