# Amazon Nova Lite vs Qwen2-72B

> Amazon Nova Lite is the stronger model overall, scoring 31.9 to 30.0 on the Noometry Index.

- Canonical page: https://noometry.com/compare/amazon-nova-lite-vs-qwen2-72b
- Last updated: 2026-10-12
- Shared benchmarks: 18

## Summary

- They share 18 benchmarks with published results for both. Amazon Nova Lite scores higher in 5 categories and Qwen2-72B in 3 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Amazon Nova Lite leads 26.3 to 21.2.
- Qwen2-72B has downloadable open weights; the other is API-only.

## Snapshot

| | Amazon Nova Lite | Qwen2-72B |
|---|---|---|
| Provider | Amazon | Alibaba (Qwen) |
| Noometry Index | 31.9 | 30.0 |
| Rank | 265 | 300 |
| Context | 300K | — |
| Input $/M | $0.06 | — |
| Output $/M | $0.24 | — |
| Weights | Proprietary | Open |

## Coding

- Amazon Nova Lite: 32.5 (#270)
- Qwen2-72B: 29.1 (#310)

| Benchmark | Amazon Nova Lite | Qwen2-72B |
|---|---|---|
| LMArena Coding | 1239 | 1196 |
| WeirdML | — | 11.3% |
| BigCodeBench Instruct | — | 38.5% |
| LiveBench Coding | 27.5% | — |
| BigCodeBench Complete | — | 54% |
| ALE-Bench | 236.25 | — |

## Agentic & Tool Use

- Amazon Nova Lite: —
- Qwen2-72B: 17.0 (#146)

| Benchmark | Amazon Nova Lite | Qwen2-72B |
|---|---|---|
| TheAgentCompany | — | 1.1% |
| METR Time Horizons | — | 29.9% |

## Reasoning

- Amazon Nova Lite: 19.3 (#260)
- Qwen2-72B: 23.2 (#181)

| Benchmark | Amazon Nova Lite | Qwen2-72B |
|---|---|---|
| LMArena Hard Prompts | 1220 | 1191 |
| LiveBench Reasoning | 36.7% | — |
| LiveBench Data Analysis | 37.2% | — |
| Epoch Capabilities Index | — | 125.28 |
| LiveBench | 36.4% | — |

## Math

- Amazon Nova Lite: 27.8 (#247)
- Qwen2-72B: 30.2 (#236)

| Benchmark | Amazon Nova Lite | Qwen2-72B |
|---|---|---|
| LMArena Math | 1227 | 1235 |
| Omni-MATH | 23.3% | — |
| LiveBench Math | 36.7% | — |
| MATH Level 5 | — | 39.1% |

## Knowledge

- Amazon Nova Lite: 26.3 (#259)
- Qwen2-72B: 21.2 (#275)

| Benchmark | Amazon Nova Lite | Qwen2-72B |
|---|---|---|
| LMArena Expert | 1201 | 1171 |
| MMLU | 77% | 82.4% |
| GPQA Diamond | — | 40.8% |
| Humanity's Last Exam | 3.6% | — |
| MMLU-Pro | 60% | — |
| Vectara Hallucination Rate | 6.1% | — |
| GPQA (HELM) | 39.7% | — |

## Multimodal

- Amazon Nova Lite: 25.5 (#123)
- Qwen2-72B: —

| Benchmark | Amazon Nova Lite | Qwen2-72B |
|---|---|---|
| LMArena Vision | 990 | — |

## Multilingual

- Amazon Nova Lite: 38.0 (#232)
- Qwen2-72B: 35.9 (#244)

| Benchmark | Amazon Nova Lite | Qwen2-72B |
|---|---|---|
| LMArena Non-English | 1208 | 1176 |
| LMArena Chinese | 1225 | 1240 |
| LMArena French | 1237 | 1170 |
| LMArena German | 1229 | 1151 |
| LMArena Japanese | 1153 | 1111 |
| LMArena Korean | 1153 | 1083 |
| LMArena Russian | 1216 | 1169 |
| LMArena Spanish | 1226 | 1169 |

## Instruction Following

- Amazon Nova Lite: 59.3 (#255)
- Qwen2-72B: 61.7 (#241)

| Benchmark | Amazon Nova Lite | Qwen2-72B |
|---|---|---|
| LMArena Instruction Following | 1205 | 1181 |
| LiveBench Instruction Following | 54.1% | — |
| IFEval | 77.6% | — |

## Long Context

- Amazon Nova Lite: 37.4 (#217)
- Qwen2-72B: 36.1 (#235)

| Benchmark | Amazon Nova Lite | Qwen2-72B |
|---|---|---|
| LMArena Longer Query | 1234 | 1192 |

## Writing & Preference

- Amazon Nova Lite: 42.1 (#237)
- Qwen2-72B: 40.8 (#241)

| Benchmark | Amazon Nova Lite | Qwen2-72B |
|---|---|---|
| LMArena Text | 1229 | 1203 |
| LMArena Creative Writing | 1197 | 1181 |
| LMArena Multi-Turn | 1199 | 1196 |
| WildBench | 75% | — |
| LiveBench Language | 25.9% | — |

## FAQ

### Is Amazon Nova Lite better than Qwen2-72B?

Amazon Nova Lite is the stronger model overall, scoring 31.9 to 30.0 on the Noometry Index.

### Is Amazon Nova Lite or Qwen2-72B better for coding?

Amazon Nova Lite scores higher on coding benchmarks: 32.5 versus 29.1 in the Noometry coding category.

### How many benchmarks do Amazon Nova Lite and Qwen2-72B share?

18 benchmarks have published results for both models. Amazon Nova Lite has 34 scored results on Noometry and Qwen2-72B has 26.