# GPT-5.4 nano vs Grok 3

> GPT-5.4 nano is the stronger model overall, scoring 41.9 to 39.9 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gpt-5-4-nano-vs-grok-3
- Last updated: 2026-10-12
- Shared benchmarks: 27

## Summary

- They share 27 benchmarks with published results for both. GPT-5.4 nano scores higher in 4 categories and Grok 3 in 4 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GPT-5.4 nano leads 23.7 to 13.7.
- The biggest single-benchmark swing is ARC-AGI-1: 51.5% for GPT-5.4 nano and 5.5% for Grok 3.

## Snapshot

| | GPT-5.4 nano | Grok 3 |
|---|---|---|
| Provider | OpenAI | xAI |
| Noometry Index | 41.9 | 39.9 |
| Rank | 125 | 157 |
| Context | 400K | — |
| Input $/M | $0.20 | — |
| Output $/M | $1.25 | — |
| Weights | Proprietary | Proprietary |

## Coding

- GPT-5.4 nano: 43.6 (#84)
- Grok 3: 41.9 (#115)

| Benchmark | GPT-5.4 nano | Grok 3 |
|---|---|---|
| WeirdML | 49.2% | 37.2% |
| LMArena Coding | 1405 | 1432 |
| Aider Polyglot | — | 53.3% |
| SciCode | 46.9% | — |
| ALE-Bench | 1,005 | — |

## Agentic & Tool Use

- GPT-5.4 nano: —
- Grok 3: 30.5 (#76)

| Benchmark | GPT-5.4 nano | Grok 3 |
|---|---|---|
| BALROG | — | 29.5% |

## Reasoning

- GPT-5.4 nano: 23.7 (#173)
- Grok 3: 13.7 (#333)

| Benchmark | GPT-5.4 nano | Grok 3 |
|---|---|---|
| ARC-AGI-2 | 5.7% | 0% |
| Kagi LLM Benchmark | 39.7% | 61.3% |
| ARC-AGI-1 | 51.5% | 5.5% |
| LMArena Hard Prompts | 1381 | 1434 |
| Epoch Capabilities Index | 145.81 | 138.33 |
| SimpleBench | — | 36.1% |
| CritPt | 9.3% | — |
| Chess Puzzles | 30% | — |
| Mystery Game Puzzles | 9% | — |
| DTBench | 80.3% | — |
| LMCA | 36.9% | — |
| ForecastBench | 57.3 | — |

## Math

- GPT-5.4 nano: 40.9 (#88)
- Grok 3: 38.0 (#145)

| Benchmark | GPT-5.4 nano | Grok 3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 87.8% | 55.6% |
| LMArena Math | 1406 | 1391 |
| FrontierMath (Feb 2025 set) | 25.9% | 3.8% |
| FrontierMath Tier 4 (v1) | 6.3% | 0% |
| FrontierMath (Tiers 1-3) | 44.9% | — |
| FrontierMath Tier 4 | 12.2% | — |
| ProofBench | 5% | — |
| Omni-MATH | — | 46.4% |
| MATH Level 5 | — | 88.7% |

## Knowledge

- GPT-5.4 nano: 41.9 (#103)
- Grok 3: 46.2 (#82)

| Benchmark | GPT-5.4 nano | Grok 3 |
|---|---|---|
| GPQA Diamond | 78.5% | 75.8% |
| Vectara Hallucination Rate | 3.1% | 5.8% |
| LMArena Expert | 1396 | 1421 |
| SimpleQA Verified | 11.7% | — |
| MMLU-Pro | — | 78.8% |
| Confabulations | — | 14.2% |
| GPQA (HELM) | — | 65% |

## Multimodal

- GPT-5.4 nano: 36.7 (#78)
- Grok 3: —

| Benchmark | GPT-5.4 nano | Grok 3 |
|---|---|---|
| LMArena Vision | 1196 | — |

## Multilingual

- GPT-5.4 nano: 48.6 (#140)
- Grok 3: 52.3 (#87)

| Benchmark | GPT-5.4 nano | Grok 3 |
|---|---|---|
| LMArena Non-English | 1359 | 1410 |
| LMArena Chinese | 1392 | 1448 |
| LMArena French | 1396 | 1460 |
| LMArena German | 1367 | 1431 |
| LMArena Japanese | 1343 | 1387 |
| LMArena Korean | 1320 | 1373 |
| LMArena Russian | 1363 | 1416 |
| LMArena Spanish | 1371 | 1417 |

## Instruction Following

- GPT-5.4 nano: 71.9 (#144)
- Grok 3: 75.0 (#73)

| Benchmark | GPT-5.4 nano | Grok 3 |
|---|---|---|
| LMArena Instruction Following | 1362 | 1409 |
| IFEval | — | 88.4% |

## Long Context

- GPT-5.4 nano: 41.6 (#137)
- Grok 3: 38.7 (#192)

| Benchmark | GPT-5.4 nano | Grok 3 |
|---|---|---|
| LMArena Longer Query | 1366 | 1439 |
| Fiction.LiveBench | — | 58.3% |

## Writing & Preference

- GPT-5.4 nano: 55.7 (#142)
- Grok 3: 55.8 (#141)

| Benchmark | GPT-5.4 nano | Grok 3 |
|---|---|---|
| LMArena Text | 1372 | 1426 |
| LMArena Creative Writing | 1314 | 1414 |
| LMArena Multi-Turn | 1382 | 1425 |
| Short-Story Creative Writing | — | 76.4% |
| EQ-Bench Creative Writing | — | 1186 |
| WildBench | — | 84.9% |

## FAQ

### Is GPT-5.4 nano better than Grok 3?

GPT-5.4 nano is the stronger model overall, scoring 41.9 to 39.9 on the Noometry Index.

### Is GPT-5.4 nano or Grok 3 better for coding?

GPT-5.4 nano scores higher on coding benchmarks: 43.6 versus 41.9 in the Noometry coding category.

### How many benchmarks do GPT-5.4 nano and Grok 3 share?

27 benchmarks have published results for both models. GPT-5.4 nano has 40 scored results on Noometry and Grok 3 has 40.