# GPT-5 Nano vs Grok-3 mini

> Grok-3 mini is the stronger model overall, scoring 41.0 to 33.4 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gpt-5-nano-vs-grok-3-mini
- Last updated: 2026-10-10
- Shared benchmarks: 30

## Summary

- They share 30 benchmarks with published results for both. GPT-5 Nano scores higher in 1 category and Grok-3 mini in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Grok-3 mini leads 52.7 to 39.4.
- The biggest single-benchmark swing is Omni-MATH: 54.6% for GPT-5 Nano and 31.8% for Grok-3 mini.

## Snapshot

| | GPT-5 Nano | Grok-3 mini |
|---|---|---|
| Provider | OpenAI | xAI |
| Noometry Index | 33.4 | 41.0 |
| Rank | 234 | 130 |
| Context | 400K | — |
| Input $/M | $0.05 | — |
| Output $/M | $0.40 | — |
| Weights | Proprietary | Proprietary |

## Coding

- GPT-5 Nano: 33.3 (#248)
- Grok-3 mini: 40.4 (#126)

| Benchmark | GPT-5 Nano | Grok-3 mini |
|---|---|---|
| WeirdML | 38.1% | 42.6% |
| LMArena Coding | 1351 | 1380 |
| SWE-bench Verified (bash only) | 34.8% | — |
| Aider Polyglot | — | 49.3% |
| ALE-Bench | 718.67 | — |

## Agentic & Tool Use

- GPT-5 Nano: 25.6 (#99)
- Grok-3 mini: —

| Benchmark | GPT-5 Nano | Grok-3 mini |
|---|---|---|
| Terminal-Bench | 21.8% | — |
| Berkeley Function Calling Leaderboard | 51.5% | — |

## Reasoning

- GPT-5 Nano: 16.4 (#291)
- Grok-3 mini: 13.8 (#323)

| Benchmark | GPT-5 Nano | Grok-3 mini |
|---|---|---|
| ARC-AGI-2 | 2.6% | 0.4% |
| Kagi LLM Benchmark | 62.2% | 61.3% |
| ARC-AGI-1 | 20.7% | 16.5% |
| LMArena Hard Prompts | 1329 | 1374 |
| Epoch Capabilities Index | 139.39 | 140.35 |
| Chess Puzzles | 27% | — |
| Mystery Game Puzzles | 9% | — |
| DTBench | 62.7% | — |
| LMCA | 7.9% | — |
| ForecastBench | 59.1 | — |

## Math

- GPT-5 Nano: 29.1 (#221)
- Grok-3 mini: 40.3 (#78)

| Benchmark | GPT-5 Nano | Grok-3 mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 81.1% | 77.8% |
| Omni-MATH | 54.6% | 31.8% |
| LMArena Math | 1318 | 1386 |
| MATH Level 5 | 95.2% | 90.9% |
| FrontierMath (Feb 2025 set) | 8.3% | 5.9% |
| FrontierMath (Tiers 1-3) | 20% | — |
| FrontierMath Tier 4 | 2.4% | — |
| ProofBench | 12% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |

## Knowledge

- GPT-5 Nano: 35.8 (#166)
- Grok-3 mini: 46.3 (#75)

| Benchmark | GPT-5 Nano | Grok-3 mini |
|---|---|---|
| GPQA Diamond | 69.4% | 76.3% |
| MMLU-Pro | 77.8% | 79.9% |
| GPQA (HELM) | 67.9% | 67.5% |
| LMArena Expert | 1321 | 1395 |
| SimpleQA Verified | 11.7% | — |
| Confabulations | — | 10.8% |
| Vectara Hallucination Rate | 10.5% | — |

## Multimodal

- GPT-5 Nano: 30.7 (#102)
- Grok-3 mini: —

| Benchmark | GPT-5 Nano | Grok-3 mini |
|---|---|---|
| LMArena Vision | 1159 | — |
| VPCT | 37.2% | — |

## Multilingual

- GPT-5 Nano: 45.5 (#167)
- Grok-3 mini: 48.3 (#141)

| Benchmark | GPT-5 Nano | Grok-3 mini |
|---|---|---|
| LMArena Non-English | 1313 | 1352 |
| LMArena Chinese | 1355 | 1387 |
| LMArena German | 1329 | 1350 |
| LMArena Korean | 1269 | 1336 |
| LMArena Russian | 1296 | 1352 |
| LMArena Spanish | 1360 | 1379 |
| LMArena French | — | 1357 |
| LMArena Japanese | — | 1343 |

## Instruction Following

- GPT-5 Nano: 75.2 (#72)
- Grok-3 mini: 78.7 (#8)

| Benchmark | GPT-5 Nano | Grok-3 mini |
|---|---|---|
| IFEval | 93.2% | 95.1% |
| LMArena Instruction Following | 1306 | 1356 |

## Long Context

- GPT-5 Nano: 31.3 (#275)
- Grok-3 mini: 41.0 (#142)

| Benchmark | GPT-5 Nano | Grok-3 mini |
|---|---|---|
| Fiction.LiveBench | 44.4% | 66.7% |
| LMArena Longer Query | 1313 | 1372 |

## Writing & Preference

- GPT-5 Nano: 39.4 (#244)
- Grok-3 mini: 52.7 (#163)

| Benchmark | GPT-5 Nano | Grok-3 mini |
|---|---|---|
| LMArena Text | 1320 | 1370 |
| LMArena Creative Writing | 1250 | 1342 |
| WildBench | 80.6% | 65.1% |
| LMArena Multi-Turn | 1311 | 1355 |
| Short-Story Creative Writing | — | 73.5% |
| EQ-Bench Creative Writing | 705 | — |

## FAQ

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

Grok-3 mini is the stronger model overall, scoring 41.0 to 33.4 on the Noometry Index.

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

Grok-3 mini scores higher on coding benchmarks: 40.4 versus 33.3 in the Noometry coding category.

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

30 benchmarks have published results for both models. GPT-5 Nano has 48 scored results on Noometry and Grok-3 mini has 35.