# GPT-5.4 mini vs Grok 4

> Grok 4 is the stronger model overall, scoring 48.1 to 45.0 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gpt-5-4-mini-vs-grok-4
- Last updated: 2026-10-10
- Shared benchmarks: 30

## Summary

- They share 30 benchmarks with published results for both. GPT-5.4 mini scores higher in 3 categories and Grok 4 in 7 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in long context, where Grok 4 leads 63.1 to 43.0.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 37.9% for GPT-5.4 mini and 73.6% for Grok 4.

## Snapshot

| | GPT-5.4 mini | Grok 4 |
|---|---|---|
| Provider | OpenAI | xAI |
| Noometry Index | 45.0 | 48.1 |
| Rank | 76 | 56 |
| Context | 400K | — |
| Input $/M | $0.75 | — |
| Output $/M | $4.50 | — |
| Weights | Proprietary | Proprietary |

## Coding

- GPT-5.4 mini: 45.2 (#72)
- Grok 4: 50.3 (#46)

| Benchmark | GPT-5.4 mini | Grok 4 |
|---|---|---|
| WeirdML | 60.3% | 45.7% |
| LMArena Coding | 1438 | 1408 |
| FrontierCode | 27% | — |
| Aider Polyglot | — | 79.6% |
| LMArena WebDev | 1397 | — |
| SciCode | 49.9% | — |
| ALE-Bench | 1,189 | — |

## Agentic & Tool Use

- GPT-5.4 mini: 29.9 (#81)
- Grok 4: 32.3 (#68)

| Benchmark | GPT-5.4 mini | Grok 4 |
|---|---|---|
| DeepResearch Bench | 36.3% | 47.3% |
| Terminal-Bench | — | 27.2% |
| Berkeley Function Calling Leaderboard | — | 63% |
| GDPval | — | 21.1% |
| Cybench | — | 43% |
| BALROG | — | 43.6% |
| LMArena Search | — | 1142 |
| METR Time Horizons | — | 66.6% |

## Reasoning

- GPT-5.4 mini: 30.4 (#85)
- Grok 4: 36.7 (#65)

| Benchmark | GPT-5.4 mini | Grok 4 |
|---|---|---|
| ARC-AGI-2 | 18.9% | 16% |
| Kagi LLM Benchmark | 37.9% | 73.6% |
| ARC-AGI-1 | 63.7% | 66.7% |
| Chess Puzzles | 24% | 28% |
| LMArena Hard Prompts | 1424 | 1409 |
| Epoch Capabilities Index | 148.84 | 146.44 |
| ForecastBench | 57 | 60.9 |
| SimpleBench | — | 60.5% |
| NYT Connections (extended) | 61.8% | — |
| CritPt | 10% | — |
| Thematic Generalization | 61.7% | — |
| Mystery Game Puzzles | 11% | — |
| DTBench | 80% | — |
| LMCA | 40.8% | — |

## Math

- GPT-5.4 mini: 45.5 (#75)
- Grok 4: 48.4 (#64)

| Benchmark | GPT-5.4 mini | Grok 4 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 88.9% | 84% |
| LMArena Math | 1419 | 1422 |
| FrontierMath (Feb 2025 set) | 28.3% | 19.7% |
| FrontierMath Tier 4 (v1) | 2.1% | 2.1% |
| FrontierMath (Tiers 1-3) | 51.2% | — |
| FrontierMath Tier 4 | 9.8% | — |
| ProofBench | 21% | — |
| Omni-MATH | — | 60.3% |

## Knowledge

- GPT-5.4 mini: 51.5 (#67)
- Grok 4: 53.8 (#55)

| Benchmark | GPT-5.4 mini | Grok 4 |
|---|---|---|
| GPQA Diamond | 86.9% | 87% |
| LMArena Expert | 1435 | 1415 |
| SimpleQA Verified | 29.4% | — |
| MMLU-Pro | — | 85.1% |
| Confabulations | — | 12.4% |
| Vectara Hallucination Rate | 5.5% | — |
| GPQA (HELM) | — | 72.7% |

## Multimodal

- GPT-5.4 mini: 39.7 (#56)
- Grok 4: 33.7 (#94)

| Benchmark | GPT-5.4 mini | Grok 4 |
|---|---|---|
| LMArena Vision | 1245 | 1210 |
| GeoBench | — | 45% |

## Multilingual

- GPT-5.4 mini: 51.9 (#96)
- Grok 4: 51.8 (#103)

| Benchmark | GPT-5.4 mini | Grok 4 |
|---|---|---|
| LMArena Non-English | 1405 | 1403 |
| LMArena Chinese | 1446 | 1427 |
| LMArena French | 1440 | 1418 |
| LMArena German | 1409 | 1429 |
| LMArena Japanese | 1374 | 1394 |
| LMArena Korean | 1368 | 1377 |
| LMArena Russian | 1417 | 1410 |
| LMArena Spanish | 1405 | 1420 |

## Instruction Following

- GPT-5.4 mini: 74.1 (#102)
- Grok 4: 79.2 (#5)

| Benchmark | GPT-5.4 mini | Grok 4 |
|---|---|---|
| LMArena Instruction Following | 1405 | 1387 |
| IFEval | — | 94.9% |

## Long Context

- GPT-5.4 mini: 43.0 (#112)
- Grok 4: 63.1 (#4)

| Benchmark | GPT-5.4 mini | Grok 4 |
|---|---|---|
| LMArena Longer Query | 1407 | 1409 |
| Fiction.LiveBench | — | 94.4% |

## Writing & Preference

- GPT-5.4 mini: 64.0 (#58)
- Grok 4: 58.5 (#116)

| Benchmark | GPT-5.4 mini | Grok 4 |
|---|---|---|
| LMArena Text | 1412 | 1411 |
| LMArena Creative Writing | 1370 | 1397 |
| LMArena Multi-Turn | 1429 | 1416 |
| Short-Story Creative Writing | — | 76.9% |
| EQ-Bench Creative Writing | 1665 | — |
| WildBench | — | 79.7% |

## FAQ

### Is GPT-5.4 mini better than Grok 4?

Grok 4 is the stronger model overall, scoring 48.1 to 45.0 on the Noometry Index.

### Is GPT-5.4 mini or Grok 4 better for coding?

Grok 4 scores higher on coding benchmarks: 50.3 versus 45.2 in the Noometry coding category.

### How many benchmarks do GPT-5.4 mini and Grok 4 share?

30 benchmarks have published results for both models. GPT-5.4 mini has 46 scored results on Noometry and Grok 4 has 48.