# Amazon Nova Lite vs Claude 3 Haiku

> Amazon Nova Lite is the stronger model overall, scoring 31.9 to 25.9 on the Noometry Index.

- Canonical page: https://noometry.com/compare/amazon-nova-lite-vs-claude-3-haiku
- Last updated: 2026-10-10
- Shared benchmarks: 19

## Summary

- They share 19 benchmarks with published results for both. Amazon Nova Lite scores higher in 8 categories and Claude 3 Haiku in 1 category; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Amazon Nova Lite leads 27.8 to 9.8.

## Snapshot

| | Amazon Nova Lite | Claude 3 Haiku |
|---|---|---|
| Provider | Amazon | Anthropic |
| Noometry Index | 31.9 | 25.9 |
| Rank | 265 | 340 |
| Context | 300K | — |
| Input $/M | $0.06 | — |
| Output $/M | $0.24 | — |
| Weights | Proprietary | Proprietary |

## Coding

- Amazon Nova Lite: 32.5 (#270)
- Claude 3 Haiku: 26.4 (#325)

| Benchmark | Amazon Nova Lite | Claude 3 Haiku |
|---|---|---|
| LMArena Coding | 1239 | 1199 |
| WeirdML | — | 9.8% |
| BigCodeBench Instruct | — | 39.4% |
| LiveBench Coding | 27.5% | — |
| BigCodeBench Complete | — | 50.1% |
| CadEval | — | 12% |
| ALE-Bench | 236.25 | — |
| HumanEval+ | — | 68.9% |
| MBPP+ | — | 68.8% |

## Reasoning

- Amazon Nova Lite: 19.3 (#260)
- Claude 3 Haiku: 16.3 (#307)

| Benchmark | Amazon Nova Lite | Claude 3 Haiku |
|---|---|---|
| LMArena Hard Prompts | 1220 | 1174 |
| Kagi LLM Benchmark | — | 34.2% |
| LiveBench Reasoning | 36.7% | — |
| DTBench | — | 50.1% |
| LiveBench Data Analysis | 37.2% | — |
| LMCA | — | 8.8% |
| Epoch Capabilities Index | — | 118.35 |
| ForecastBench | — | 53.2 |
| LiveBench | 36.4% | — |
| WinoGrande | — | 74.2% |

## Math

- Amazon Nova Lite: 27.8 (#247)
- Claude 3 Haiku: 9.8 (#319)

| Benchmark | Amazon Nova Lite | Claude 3 Haiku |
|---|---|---|
| LMArena Math | 1227 | 1188 |
| OTIS Mock AIME 2024-2025 | — | 1.8% |
| Omni-MATH | 23.3% | — |
| LiveBench Math | 36.7% | — |
| MATH Level 5 | — | 14.9% |

## Knowledge

- Amazon Nova Lite: 26.3 (#259)
- Claude 3 Haiku: 17.3 (#285)

| Benchmark | Amazon Nova Lite | Claude 3 Haiku |
|---|---|---|
| LMArena Expert | 1201 | 1148 |
| MMLU | 77% | 73.8% |
| GPQA Diamond | — | 36.3% |
| Humanity's Last Exam | 3.6% | — |
| MMLU-Pro | 60% | — |
| Confabulations | — | 34.2% |
| Vectara Hallucination Rate | 6.1% | — |
| GPQA (HELM) | 39.7% | — |

## Multimodal

- Amazon Nova Lite: 25.5 (#123)
- Claude 3 Haiku: 23.6 (#128)

| Benchmark | Amazon Nova Lite | Claude 3 Haiku |
|---|---|---|
| LMArena Vision | 990 | 950 |
| ScienceQA | — | 72% |

## Multilingual

- Amazon Nova Lite: 38.0 (#232)
- Claude 3 Haiku: 36.0 (#243)

| Benchmark | Amazon Nova Lite | Claude 3 Haiku |
|---|---|---|
| LMArena Non-English | 1208 | 1178 |
| LMArena Chinese | 1225 | 1155 |
| LMArena French | 1237 | 1195 |
| LMArena German | 1229 | 1174 |
| LMArena Japanese | 1153 | 1102 |
| LMArena Korean | 1153 | 1109 |
| LMArena Russian | 1216 | 1204 |
| LMArena Spanish | 1226 | 1166 |

## Instruction Following

- Amazon Nova Lite: 59.3 (#255)
- Claude 3 Haiku: 61.3 (#247)

| Benchmark | Amazon Nova Lite | Claude 3 Haiku |
|---|---|---|
| LMArena Instruction Following | 1205 | 1173 |
| LiveBench Instruction Following | 54.1% | — |
| IFEval | 77.6% | — |

## Long Context

- Amazon Nova Lite: 37.4 (#217)
- Claude 3 Haiku: 36.1 (#237)

| Benchmark | Amazon Nova Lite | Claude 3 Haiku |
|---|---|---|
| LMArena Longer Query | 1234 | 1190 |

## Writing & Preference

- Amazon Nova Lite: 42.1 (#237)
- Claude 3 Haiku: 29.7 (#291)

| Benchmark | Amazon Nova Lite | Claude 3 Haiku |
|---|---|---|
| LMArena Text | 1229 | 1195 |
| LMArena Creative Writing | 1197 | 1157 |
| LMArena Multi-Turn | 1199 | 1190 |
| EQ-Bench Creative Writing | — | 717 |
| WildBench | 75% | — |
| LiveBench Language | 25.9% | — |

## FAQ

### Is Amazon Nova Lite better than Claude 3 Haiku?

Amazon Nova Lite is the stronger model overall, scoring 31.9 to 25.9 on the Noometry Index.

### Is Amazon Nova Lite or Claude 3 Haiku better for coding?

Amazon Nova Lite scores higher on coding benchmarks: 32.5 versus 26.4 in the Noometry coding category.

### How many benchmarks do Amazon Nova Lite and Claude 3 Haiku share?

19 benchmarks have published results for both models. Amazon Nova Lite has 34 scored results on Noometry and Claude 3 Haiku has 37.