# Amazon Nova Pro vs GPT-4o mini

> Amazon Nova Pro is the stronger model overall, scoring 31.0 to 25.5 on the Noometry Index. GPT-4o mini costs 5.3× less per token, which makes it the better buy when Amazon Nova Pro's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/amazon-nova-pro-vs-gpt-4o-mini
- Last updated: 2026-10-11
- Shared benchmarks: 34

## Summary

- They share 34 benchmarks with published results for both. Amazon Nova Pro scores higher in 6 categories and GPT-4o mini in 4 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Amazon Nova Pro leads 28.5 to 10.4.
- The biggest single-benchmark swing is LiveBench Instruction Following: 67.1% for Amazon Nova Pro and 56.8% for GPT-4o mini.
- GPT-4o mini is cheaper at $0.15 / $0.60 per million input/output tokens, against $0.80 / $3.20 for Amazon Nova Pro.
- Amazon Nova Pro accepts more context: 300K tokens versus 128K.

## Snapshot

| | Amazon Nova Pro | GPT-4o mini |
|---|---|---|
| Provider | Amazon | OpenAI |
| Noometry Index | 31.0 | 25.5 |
| Rank | 281 | 343 |
| Context | 300K | 128K |
| Input $/M | $0.80 | $0.15 |
| Output $/M | $3.20 | $0.60 |
| Weights | Proprietary | Proprietary |

## Coding

- Amazon Nova Pro: 35.1 (#229)
- GPT-4o mini: 22.0 (#335)

| Benchmark | Amazon Nova Pro | GPT-4o mini |
|---|---|---|
| LiveBench Coding | 38.1% | 43.1% |
| LMArena Coding | 1270 | 1290 |
| Aider Polyglot | — | 3.6% |
| WeirdML | — | 11.8% |
| BigCodeBench Instruct | — | 46.1% |
| BigCodeBench Complete | — | 57.4% |
| HumanEval+ | — | 83.5% |
| MBPP+ | — | 72.2% |

## Agentic & Tool Use

- Amazon Nova Pro: 16.7 (#147)
- GPT-4o mini: 27.5 (#101)

| Benchmark | Amazon Nova Pro | GPT-4o mini |
|---|---|---|
| Berkeley Function Calling Leaderboard | 25% | — |
| TheAgentCompany | 1.7% | — |
| BALROG | — | 17.4% |

## Reasoning

- Amazon Nova Pro: 20.0 (#243)
- GPT-4o mini: 8.7 (#347)

| Benchmark | Amazon Nova Pro | GPT-4o mini |
|---|---|---|
| LiveBench Reasoning | 32.6% | 32.8% |
| LMArena Hard Prompts | 1246 | 1267 |
| LiveBench Data Analysis | 48.3% | 50% |
| Epoch Capabilities Index | 123.8 | 126.56 |
| LiveBench | 43.5% | 41.3% |
| ARC-AGI-2 | — | 0% |
| SimpleBench | — | 10.7% |
| Kagi LLM Benchmark | — | 28.8% |
| Chess Puzzles | — | 0% |
| Mystery Game Puzzles | — | 12% |
| DTBench | — | 54.4% |
| LMCA | — | 10.4% |
| PIQA | — | 88.7% |

## Math

- Amazon Nova Pro: 28.5 (#243)
- GPT-4o mini: 10.4 (#314)

| Benchmark | Amazon Nova Pro | GPT-4o mini |
|---|---|---|
| Omni-MATH | 24.2% | 28% |
| LiveBench Math | 38% | 36.3% |
| LMArena Math | 1252 | 1267 |
| FrontierMath (Tiers 1-3) | — | 0.7% |
| OTIS Mock AIME 2024-2025 | — | 6.9% |
| MATH Level 5 | — | 52.6% |
| GSM8K | — | 91.3% |

## Knowledge

- Amazon Nova Pro: 27.4 (#250)
- GPT-4o mini: 17.7 (#284)

| Benchmark | Amazon Nova Pro | GPT-4o mini |
|---|---|---|
| MMLU-Pro | 67.3% | 60.3% |
| Confabulations | 30.1% | 37.2% |
| GPQA (HELM) | 44.6% | 36.8% |
| LMArena Expert | 1211 | 1235 |
| MMLU | 82% | 81.8% |
| GPQA Diamond | — | 37.7% |
| Humanity's Last Exam | 4.4% | — |
| SimpleQA Verified | — | 8.3% |
| Vectara Hallucination Rate | 5.1% | — |
| BoolQ | — | 88.7% |

## Multimodal

- Amazon Nova Pro: 25.0 (#126)
- GPT-4o mini: 25.9 (#122)

| Benchmark | Amazon Nova Pro | GPT-4o mini |
|---|---|---|
| LMArena Vision | 980 | 1066 |
| Video-MME | — | 64.8% |
| GeoBench | — | 64% |
| VPCT | — | 34% |

## Multilingual

- Amazon Nova Pro: 39.7 (#223)
- GPT-4o mini: 42.0 (#199)

| Benchmark | Amazon Nova Pro | GPT-4o mini |
|---|---|---|
| LMArena Non-English | 1234 | 1266 |
| LMArena Chinese | 1244 | 1265 |
| LMArena French | 1271 | 1297 |
| LMArena German | 1243 | 1272 |
| LMArena Japanese | 1200 | 1216 |
| LMArena Korean | 1203 | 1195 |
| LMArena Russian | 1240 | 1275 |
| LMArena Spanish | 1182 | 1276 |

## Instruction Following

- Amazon Nova Pro: 64.9 (#226)
- GPT-4o mini: 61.9 (#239)

| Benchmark | Amazon Nova Pro | GPT-4o mini |
|---|---|---|
| LiveBench Instruction Following | 67.1% | 56.8% |
| IFEval | 81.5% | 78.2% |
| LMArena Instruction Following | 1235 | 1258 |

## Long Context

- Amazon Nova Pro: 38.1 (#205)
- GPT-4o mini: 39.1 (#186)

| Benchmark | Amazon Nova Pro | GPT-4o mini |
|---|---|---|
| LMArena Longer Query | 1255 | 1289 |

## Writing & Preference

- Amazon Nova Pro: 43.9 (#226)
- GPT-4o mini: 39.5 (#248)

| Benchmark | Amazon Nova Pro | GPT-4o mini |
|---|---|---|
| LMArena Text | 1259 | 1286 |
| LMArena Creative Writing | 1212 | 1268 |
| Short-Story Creative Writing | 60.5% | 67.2% |
| WildBench | 77.7% | 79.1% |
| LMArena Multi-Turn | 1246 | 1285 |
| LiveBench Language | 37% | 28.6% |
| EQ-Bench Creative Writing | — | 873 |

## FAQ

### Is Amazon Nova Pro better than GPT-4o mini?

Amazon Nova Pro is the stronger model overall, scoring 31.0 to 25.5 on the Noometry Index. GPT-4o mini costs 5.3× less per token, which makes it the better buy when Amazon Nova Pro's lead doesn't matter for your workload.

### Which is cheaper, Amazon Nova Pro or GPT-4o mini?

GPT-4o mini is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; Amazon Nova Pro lists at $0.80 and $3.20.

### Is Amazon Nova Pro or GPT-4o mini better for coding?

Amazon Nova Pro scores higher on coding benchmarks: 35.1 versus 22.0 in the Noometry coding category.

### Which has the bigger context window?

Amazon Nova Pro does, with 300K tokens against 128K.

### How many benchmarks do Amazon Nova Pro and GPT-4o mini share?

34 benchmarks have published results for both models. Amazon Nova Pro has 38 scored results on Noometry and GPT-4o mini has 60.