# GPT-5 Mini vs Grok-3 mini

> GPT-5 Mini and Grok-3 mini score almost the same on the Noometry Index (41.8 vs 41.2), so choose on price, context window or the category you care about most.

- Canonical page: https://noometry.com/compare/gpt-5-mini-vs-grok-3-mini
- Last updated: 2026-10-12
- Shared benchmarks: 34

## Summary

- They share 34 benchmarks with published results for both. GPT-5 Mini scores higher in 5 categories and Grok-3 mini in 3 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GPT-5 Mini leads 23.9 to 13.6.
- The biggest single-benchmark swing is Omni-MATH: 72.2% for GPT-5 Mini and 31.8% for Grok-3 mini.

## Snapshot

| | GPT-5 Mini | Grok-3 mini |
|---|---|---|
| Provider | OpenAI | xAI |
| Noometry Index | 41.8 | 41.2 |
| Rank | 128 | 141 |
| Context | 400K | — |
| Input $/M | $0.25 | — |
| Output $/M | $2 | — |
| Weights | Proprietary | Proprietary |

## Coding

- GPT-5 Mini: 40.1 (#146)
- Grok-3 mini: 40.8 (#131)

| Benchmark | GPT-5 Mini | Grok-3 mini |
|---|---|---|
| WeirdML | 52.7% | 42.6% |
| LMArena Coding | 1406 | 1379 |
| SWE-bench Verified | 64.7% | — |
| SWE-bench Verified (bash only) | 59.8% | — |
| Aider Polyglot | — | 49.3% |
| SWE-bench Multilingual | 39.7% | — |
| SciCode | 39.2% | — |
| ALE-Bench | 799.77 | — |
| AlgoTune | 1.38 | — |

## Agentic & Tool Use

- GPT-5 Mini: 31.1 (#70)
- Grok-3 mini: —

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

## Reasoning

- GPT-5 Mini: 23.9 (#168)
- Grok-3 mini: 13.6 (#334)

| Benchmark | GPT-5 Mini | Grok-3 mini |
|---|---|---|
| ARC-AGI-2 | 4.4% | 0.4% |
| Kagi LLM Benchmark | 70.3% | 61.3% |
| ARC-AGI-1 | 54.3% | 16.5% |
| LMArena Hard Prompts | 1380 | 1375 |
| Epoch Capabilities Index | 145.52 | 140.35 |
| CritPt | 0% | — |
| Chess Puzzles | 30% | — |
| EnigmaEval | 8.2% | — |
| Mystery Game Puzzles | 10% | — |
| DTBench | 80.5% | — |
| LMCA | 34.2% | — |
| ForecastBench | 61 | — |

## Math

- GPT-5 Mini: 46.7 (#69)
- Grok-3 mini: 42.1 (#85)

| Benchmark | GPT-5 Mini | Grok-3 mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 86.7% | 77.8% |
| Omni-MATH | 72.2% | 31.8% |
| LMArena Math | 1378 | 1386 |
| MATH Level 5 | 97.8% | 90.9% |
| FrontierMath (Feb 2025 set) | 27.2% | 5.9% |
| FrontierMath (Tiers 1-3) | 46.7% | — |
| FrontierMath Tier 4 | 12.2% | — |
| ProofBench | 9% | — |
| FrontierMath Tier 4 (v1) | 6.3% | — |

## Knowledge

- GPT-5 Mini: 45.6 (#86)
- Grok-3 mini: 46.4 (#81)

| Benchmark | GPT-5 Mini | Grok-3 mini |
|---|---|---|
| GPQA Diamond | 75% | 76.3% |
| MMLU-Pro | 83.5% | 79.9% |
| Confabulations | 13.3% | 10.8% |
| GPQA (HELM) | 75.6% | 67.5% |
| LMArena Expert | 1379 | 1395 |
| Humanity's Last Exam | 19.4% | — |
| SimpleQA Verified | 21.6% | — |
| Vectara Hallucination Rate | 12.9% | — |

## Multimodal

- GPT-5 Mini: 35.6 (#85)
- Grok-3 mini: —

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

## Multilingual

- GPT-5 Mini: 48.9 (#137)
- Grok-3 mini: 48.1 (#145)

| Benchmark | GPT-5 Mini | Grok-3 mini |
|---|---|---|
| LMArena Non-English | 1363 | 1352 |
| LMArena Chinese | 1385 | 1387 |
| LMArena French | 1386 | 1357 |
| LMArena German | 1366 | 1349 |
| LMArena Japanese | 1341 | 1342 |
| LMArena Korean | 1308 | 1335 |
| LMArena Russian | 1362 | 1353 |
| LMArena Spanish | 1355 | 1381 |

## Instruction Following

- GPT-5 Mini: 76.2 (#46)
- Grok-3 mini: 78.5 (#9)

| Benchmark | GPT-5 Mini | Grok-3 mini |
|---|---|---|
| IFEval | 92.7% | 95.1% |
| LMArena Instruction Following | 1357 | 1357 |

## Long Context

- GPT-5 Mini: 41.9 (#132)
- Grok-3 mini: 41.0 (#147)

| Benchmark | GPT-5 Mini | Grok-3 mini |
|---|---|---|
| Fiction.LiveBench | 69.4% | 66.7% |
| LMArena Longer Query | 1355 | 1372 |

## Writing & Preference

- GPT-5 Mini: 55.2 (#148)
- Grok-3 mini: 52.5 (#169)

| Benchmark | GPT-5 Mini | Grok-3 mini |
|---|---|---|
| LMArena Text | 1373 | 1370 |
| LMArena Creative Writing | 1325 | 1342 |
| Short-Story Creative Writing | 83.1% | 73.5% |
| WildBench | 85.5% | 65.1% |
| LMArena Multi-Turn | 1363 | 1355 |
| EQ-Bench Creative Writing | 1313 | — |

## FAQ

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

GPT-5 Mini and Grok-3 mini score almost the same on the Noometry Index (41.8 vs 41.2), so choose on price, context window or the category you care about most.

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

They score almost the same on coding (40.1 vs 40.8); test both on your own repository before choosing.

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

34 benchmarks have published results for both models. GPT-5 Mini has 60 scored results on Noometry and Grok-3 mini has 35.