# Amazon Nova Pro vs GPT-4

> Amazon Nova Pro is the stronger model overall, scoring 31.0 to 29.1 on the Noometry Index.

- Canonical page: https://noometry.com/compare/amazon-nova-pro-vs-gpt-4
- Last updated: 2026-10-11
- Shared benchmarks: 19

## Summary

- They share 19 benchmarks with published results for both. Amazon Nova Pro scores higher in 6 categories and GPT-4 in 2 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in math, where Amazon Nova Pro leads 28.5 to 10.8.
- Amazon Nova Pro is cheaper at $0.80 / $3.20 per million input/output tokens, against $30 / $60 for GPT-4.
- Amazon Nova Pro accepts more context: 300K tokens versus 8K.

## Snapshot

| | Amazon Nova Pro | GPT-4 |
|---|---|---|
| Provider | Amazon | OpenAI |
| Noometry Index | 31.0 | 29.1 |
| Rank | 281 | 316 |
| Context | 300K | 8K |
| Input $/M | $0.80 | $30 |
| Output $/M | $3.20 | $60 |
| Weights | Proprietary | Proprietary |

## Coding

- Amazon Nova Pro: 35.1 (#229)
- GPT-4: 31.6 (#283)

| Benchmark | Amazon Nova Pro | GPT-4 |
|---|---|---|
| LMArena Coding | 1270 | 1254 |
| WeirdML | — | 12.4% |
| BigCodeBench Instruct | — | 46% |
| LiveBench Coding | 38.1% | — |
| BigCodeBench Complete | — | 57.2% |
| HumanEval+ | — | 79.3% |

## Agentic & Tool Use

- Amazon Nova Pro: 16.7 (#147)
- GPT-4: —

| Benchmark | Amazon Nova Pro | GPT-4 |
|---|---|---|
| Berkeley Function Calling Leaderboard | 25% | — |
| TheAgentCompany | 1.7% | — |
| METR Time Horizons | — | 36.1% |

## Reasoning

- Amazon Nova Pro: 20.0 (#243)
- GPT-4: 17.8 (#289)

| Benchmark | Amazon Nova Pro | GPT-4 |
|---|---|---|
| LMArena Hard Prompts | 1246 | 1241 |
| Epoch Capabilities Index | 123.8 | 125.89 |
| Chess Puzzles | — | 4% |
| LiveBench Reasoning | 32.6% | — |
| Mystery Game Puzzles | — | 12% |
| DTBench | — | 62.7% |
| LiveBench Data Analysis | 48.3% | — |
| LMCA | — | 17.1% |
| BIG-Bench Hard | — | 75.1% |
| ForecastBench | — | 57.8 |
| HellaSwag | — | 95.3% |
| LiveBench | 43.5% | — |
| WinoGrande | — | 87.5% |

## Math

- Amazon Nova Pro: 28.5 (#243)
- GPT-4: 10.8 (#309)

| Benchmark | Amazon Nova Pro | GPT-4 |
|---|---|---|
| LMArena Math | 1252 | 1269 |
| OTIS Mock AIME 2024-2025 | — | 1.1% |
| Omni-MATH | 24.2% | — |
| LiveBench Math | 38% | — |
| MATH Level 5 | — | 23% |
| GSM8K | — | 92% |

## Knowledge

- Amazon Nova Pro: 27.4 (#250)
- GPT-4: 18.4 (#282)

| Benchmark | Amazon Nova Pro | GPT-4 |
|---|---|---|
| LMArena Expert | 1211 | 1211 |
| MMLU | 82% | 86.4% |
| GPQA Diamond | — | 35.7% |
| Humanity's Last Exam | 4.4% | — |
| MMLU-Pro | 67.3% | — |
| Confabulations | 30.1% | — |
| Vectara Hallucination Rate | 5.1% | — |
| GPQA (HELM) | 44.6% | — |
| TriviaQA | — | 84.8% |

## Multimodal

- Amazon Nova Pro: 25.0 (#126)
- GPT-4: —

| Benchmark | Amazon Nova Pro | GPT-4 |
|---|---|---|
| LMArena Vision | 980 | — |

## Multilingual

- Amazon Nova Pro: 39.7 (#223)
- GPT-4: 40.6 (#215)

| Benchmark | Amazon Nova Pro | GPT-4 |
|---|---|---|
| LMArena Non-English | 1234 | 1246 |
| LMArena Chinese | 1244 | 1242 |
| LMArena French | 1271 | 1283 |
| LMArena German | 1243 | 1251 |
| LMArena Japanese | 1200 | 1209 |
| LMArena Korean | 1203 | 1184 |
| LMArena Russian | 1240 | 1251 |
| LMArena Spanish | 1182 | 1261 |

## Instruction Following

- Amazon Nova Pro: 64.9 (#226)
- GPT-4: 65.3 (#222)

| Benchmark | Amazon Nova Pro | GPT-4 |
|---|---|---|
| LMArena Instruction Following | 1235 | 1241 |
| LiveBench Instruction Following | 67.1% | — |
| IFEval | 81.5% | — |

## Long Context

- Amazon Nova Pro: 38.1 (#205)
- GPT-4: 37.7 (#212)

| Benchmark | Amazon Nova Pro | GPT-4 |
|---|---|---|
| LMArena Longer Query | 1255 | 1244 |

## Writing & Preference

- Amazon Nova Pro: 43.9 (#226)
- GPT-4: 34.9 (#268)

| Benchmark | Amazon Nova Pro | GPT-4 |
|---|---|---|
| LMArena Text | 1259 | 1263 |
| LMArena Creative Writing | 1212 | 1244 |
| LMArena Multi-Turn | 1246 | 1257 |
| Short-Story Creative Writing | 60.5% | — |
| EQ-Bench Creative Writing | — | 752 |
| WildBench | 77.7% | — |
| LiveBench Language | 37% | — |

## FAQ

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

Amazon Nova Pro is the stronger model overall, scoring 31.0 to 29.1 on the Noometry Index.

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

Amazon Nova Pro is cheaper. It lists at $0.80 per million input tokens and $3.20 per million output tokens; GPT-4 lists at $30 and $60.

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

Amazon Nova Pro scores higher on coding benchmarks: 35.1 versus 31.6 in the Noometry coding category.

### Which has the bigger context window?

Amazon Nova Pro does, with 300K tokens against 8K.

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

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