# GPT-5 Mini vs Grok 3

> GPT-5 Mini is the stronger model overall, scoring 41.7 to 39.6 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gpt-5-mini-vs-grok-3
- Last updated: 2026-10-10
- Shared benchmarks: 37

## Summary

- They share 37 benchmarks with published results for both. GPT-5 Mini scores higher in 5 categories and Grok 3 in 4 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GPT-5 Mini leads 24.1 to 13.8.
- The biggest single-benchmark swing is ARC-AGI-1: 54.3% for GPT-5 Mini and 5.5% for Grok 3.

## Snapshot

| | GPT-5 Mini | Grok 3 |
|---|---|---|
| Provider | OpenAI | xAI |
| Noometry Index | 41.7 | 39.6 |
| Rank | 115 | 150 |
| Context | 400K | — |
| Input $/M | $0.25 | — |
| Output $/M | $2 | — |
| Weights | Proprietary | Proprietary |

## Coding

- GPT-5 Mini: 40.0 (#134)
- Grok 3: 41.5 (#113)

| Benchmark | GPT-5 Mini | Grok 3 |
|---|---|---|
| WeirdML | 52.7% | 37.2% |
| LMArena Coding | 1406 | 1432 |
| SWE-bench Verified | 64.7% | — |
| SWE-bench Verified (bash only) | 59.8% | — |
| Aider Polyglot | — | 53.3% |
| SWE-bench Multilingual | 39.7% | — |
| SciCode | 39.2% | — |
| ALE-Bench | 799.77 | — |
| AlgoTune | 1.38 | — |

## Agentic & Tool Use

- GPT-5 Mini: 31.0 (#64)
- Grok 3: 30.2 (#69)

| Benchmark | GPT-5 Mini | Grok 3 |
|---|---|---|
| Terminal-Bench | 34.8% | — |
| Berkeley Function Calling Leaderboard | 55.5% | — |
| BALROG | — | 29.5% |
| Vending-Bench 2 | -31.18 | — |

## Reasoning

- GPT-5 Mini: 24.1 (#153)
- Grok 3: 13.8 (#322)

| Benchmark | GPT-5 Mini | Grok 3 |
|---|---|---|
| ARC-AGI-2 | 4.4% | 0% |
| Kagi LLM Benchmark | 70.3% | 61.3% |
| ARC-AGI-1 | 54.3% | 5.5% |
| LMArena Hard Prompts | 1380 | 1433 |
| Epoch Capabilities Index | 145.51 | 138.34 |
| SimpleBench | — | 36.1% |
| CritPt | 0% | — |
| Chess Puzzles | 30% | — |
| EnigmaEval | 8.2% | — |
| Mystery Game Puzzles | 10% | — |
| DTBench | 80.5% | — |
| LMCA | 34.2% | — |
| ForecastBench | 61 | — |

## Math

- GPT-5 Mini: 46.3 (#59)
- Grok 3: 36.2 (#122)

| Benchmark | GPT-5 Mini | Grok 3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 86.7% | 55.6% |
| Omni-MATH | 72.2% | 46.4% |
| LMArena Math | 1377 | 1390 |
| MATH Level 5 | 97.8% | 88.7% |
| FrontierMath (Feb 2025 set) | 27.2% | 3.8% |
| FrontierMath Tier 4 (v1) | 6.3% | 0% |
| FrontierMath (Tiers 1-3) | 46.7% | — |
| FrontierMath Tier 4 | 12.2% | — |
| ProofBench | 9% | — |

## Knowledge

- GPT-5 Mini: 45.5 (#80)
- Grok 3: 46.0 (#76)

| Benchmark | GPT-5 Mini | Grok 3 |
|---|---|---|
| GPQA Diamond | 75% | 75.8% |
| MMLU-Pro | 83.5% | 78.8% |
| Confabulations | 13.3% | 14.2% |
| Vectara Hallucination Rate | 12.9% | 5.8% |
| GPQA (HELM) | 75.6% | 65% |
| LMArena Expert | 1377 | 1420 |
| Humanity's Last Exam | 19.4% | — |
| SimpleQA Verified | 21.6% | — |

## Multimodal

- GPT-5 Mini: 35.0 (#80)
- Grok 3: —

| Benchmark | GPT-5 Mini | Grok 3 |
|---|---|---|
| LMArena Vision | 1202 | — |
| VPCT | 40.2% | — |

## Multilingual

- GPT-5 Mini: 49.1 (#131)
- Grok 3: 52.5 (#81)

| Benchmark | GPT-5 Mini | Grok 3 |
|---|---|---|
| LMArena Non-English | 1363 | 1411 |
| LMArena Chinese | 1387 | 1448 |
| LMArena French | 1385 | 1460 |
| LMArena German | 1366 | 1431 |
| LMArena Japanese | 1342 | 1389 |
| LMArena Korean | 1306 | 1374 |
| LMArena Russian | 1362 | 1416 |
| LMArena Spanish | 1355 | 1417 |

## Instruction Following

- GPT-5 Mini: 76.4 (#43)
- Grok 3: 75.2 (#69)

| Benchmark | GPT-5 Mini | Grok 3 |
|---|---|---|
| IFEval | 92.7% | 88.4% |
| LMArena Instruction Following | 1357 | 1409 |

## Long Context

- GPT-5 Mini: 41.9 (#127)
- Grok 3: 38.7 (#187)

| Benchmark | GPT-5 Mini | Grok 3 |
|---|---|---|
| Fiction.LiveBench | 69.4% | 58.3% |
| LMArena Longer Query | 1355 | 1439 |

## Writing & Preference

- GPT-5 Mini: 55.4 (#143)
- Grok 3: 56.0 (#136)

| Benchmark | GPT-5 Mini | Grok 3 |
|---|---|---|
| LMArena Text | 1373 | 1426 |
| LMArena Creative Writing | 1325 | 1414 |
| Short-Story Creative Writing | 83.1% | 76.4% |
| EQ-Bench Creative Writing | 1313 | 1186 |
| WildBench | 85.5% | 84.9% |
| LMArena Multi-Turn | 1363 | 1424 |

## FAQ

### Is GPT-5 Mini better than Grok 3?

GPT-5 Mini is the stronger model overall, scoring 41.7 to 39.6 on the Noometry Index.

### Is GPT-5 Mini or Grok 3 better for coding?

Grok 3 scores higher on coding benchmarks: 41.5 versus 40.0 in the Noometry coding category.

### How many benchmarks do GPT-5 Mini and Grok 3 share?

37 benchmarks have published results for both models. GPT-5 Mini has 60 scored results on Noometry and Grok 3 has 40.