# Amazon Nova Lite vs Llama 4 Maverick

> Amazon Nova Lite is the stronger model overall, scoring 31.9 to 30.9 on the Noometry Index.

- Canonical page: https://noometry.com/compare/amazon-nova-lite-vs-llama-4-maverick
- Last updated: 2026-10-11
- Shared benchmarks: 26

## Summary

- They share 26 benchmarks with published results for both. Amazon Nova Lite scores higher in 5 categories and Llama 4 Maverick in 4 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in instruction following, where Llama 4 Maverick leads 71.7 to 59.3.
- The biggest single-benchmark swing is GPQA (HELM): 39.7% for Amazon Nova Lite and 65% for Llama 4 Maverick.
- Amazon Nova Lite is cheaper at $0.06 / $0.24 per million input/output tokens, against $0.19 / $0.65 for Llama 4 Maverick.
- Amazon Nova Lite accepts more context: 300K tokens versus 128K.
- Llama 4 Maverick has downloadable open weights; the other is API-only.

## Snapshot

| | Amazon Nova Lite | Llama 4 Maverick |
|---|---|---|
| Provider | Amazon | Meta |
| Noometry Index | 31.9 | 30.9 |
| Rank | 265 | 282 |
| Context | 300K | 128K |
| Input $/M | $0.06 | $0.19 |
| Output $/M | $0.24 | $0.65 |
| Weights | Proprietary | Open |

## Coding

- Amazon Nova Lite: 32.5 (#270)
- Llama 4 Maverick: 26.6 (#324)

| Benchmark | Amazon Nova Lite | Llama 4 Maverick |
|---|---|---|
| LMArena Coding | 1239 | 1302 |
| ALE-Bench | 236.25 | 172.97 |
| SWE-bench Verified (bash only) | — | 21% |
| Aider Polyglot | — | 15.6% |
| SciCode | — | 33.1% |
| WeirdML | — | 24.5% |
| BigCodeBench Instruct | — | 49.7% |
| LiveBench Coding | 27.5% | — |
| BigCodeBench Complete | — | 61.4% |

## Agentic & Tool Use

- Amazon Nova Lite: —
- Llama 4 Maverick: 28.2 (#91)

| Benchmark | Amazon Nova Lite | Llama 4 Maverick |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 37.3% |

## Reasoning

- Amazon Nova Lite: 19.3 (#260)
- Llama 4 Maverick: 10.1 (#342)

| Benchmark | Amazon Nova Lite | Llama 4 Maverick |
|---|---|---|
| LMArena Hard Prompts | 1220 | 1281 |
| ARC-AGI-2 | — | 0% |
| SimpleBench | — | 27.7% |
| Kagi LLM Benchmark | — | 55.9% |
| NYT Connections (extended) | — | 8% |
| ARC-AGI-1 | — | 4.4% |
| CritPt | — | 0% |
| EnigmaEval | — | 0.6% |
| LiveBench Reasoning | 36.7% | — |
| DTBench | — | 61.9% |
| LiveBench Data Analysis | 37.2% | — |
| LMCA | — | 15.9% |
| Epoch Capabilities Index | — | 132.2 |
| ForecastBench | — | 57.5 |
| LiveBench | 36.4% | — |

## Math

- Amazon Nova Lite: 27.8 (#247)
- Llama 4 Maverick: 26.0 (#262)

| Benchmark | Amazon Nova Lite | Llama 4 Maverick |
|---|---|---|
| Omni-MATH | 23.3% | 42.2% |
| LMArena Math | 1227 | 1299 |
| OTIS Mock AIME 2024-2025 | — | 20.6% |
| LiveBench Math | 36.7% | — |
| MATH Level 5 | — | 73% |
| FrontierMath (Feb 2025 set) | — | 0.7% |

## Knowledge

- Amazon Nova Lite: 26.3 (#259)
- Llama 4 Maverick: 33.4 (#204)

| Benchmark | Amazon Nova Lite | Llama 4 Maverick |
|---|---|---|
| Humanity's Last Exam | 3.6% | 5.7% |
| MMLU-Pro | 60% | 81% |
| Vectara Hallucination Rate | 6.1% | 8.2% |
| GPQA (HELM) | 39.7% | 65% |
| LMArena Expert | 1201 | 1259 |
| GPQA Diamond | — | 67% |
| Confabulations | — | 22.6% |
| MMLU | 77% | — |

## Multimodal

- Amazon Nova Lite: 25.5 (#123)
- Llama 4 Maverick: 31.6 (#105)

| Benchmark | Amazon Nova Lite | Llama 4 Maverick |
|---|---|---|
| LMArena Vision | 990 | 1142 |
| GeoBench | — | 52% |
| SpatialViz-Bench | — | 31.8% |

## Multilingual

- Amazon Nova Lite: 38.0 (#232)
- Llama 4 Maverick: 42.2 (#195)

| Benchmark | Amazon Nova Lite | Llama 4 Maverick |
|---|---|---|
| LMArena Non-English | 1208 | 1269 |
| LMArena Chinese | 1225 | 1277 |
| LMArena French | 1237 | 1259 |
| LMArena German | 1229 | 1291 |
| LMArena Japanese | 1153 | 1207 |
| LMArena Korean | 1153 | 1203 |
| LMArena Russian | 1216 | 1286 |
| LMArena Spanish | 1226 | 1293 |

## Instruction Following

- Amazon Nova Lite: 59.3 (#255)
- Llama 4 Maverick: 71.7 (#146)

| Benchmark | Amazon Nova Lite | Llama 4 Maverick |
|---|---|---|
| IFEval | 77.6% | 90.8% |
| LMArena Instruction Following | 1205 | 1267 |
| LiveBench Instruction Following | 54.1% | — |

## Long Context

- Amazon Nova Lite: 37.4 (#217)
- Llama 4 Maverick: 31.4 (#279)

| Benchmark | Amazon Nova Lite | Llama 4 Maverick |
|---|---|---|
| LMArena Longer Query | 1234 | 1280 |
| Fiction.LiveBench | — | 46.2% |

## Writing & Preference

- Amazon Nova Lite: 42.1 (#237)
- Llama 4 Maverick: 38.8 (#252)

| Benchmark | Amazon Nova Lite | Llama 4 Maverick |
|---|---|---|
| LMArena Text | 1229 | 1287 |
| LMArena Creative Writing | 1197 | 1267 |
| WildBench | 75% | 80% |
| LMArena Multi-Turn | 1199 | 1289 |
| Short-Story Creative Writing | — | 62% |
| EQ-Bench Creative Writing | — | 860 |
| LiveBench Language | 25.9% | — |

## FAQ

### Is Amazon Nova Lite better than Llama 4 Maverick?

Amazon Nova Lite is the stronger model overall, scoring 31.9 to 30.9 on the Noometry Index.

### Which is cheaper, Amazon Nova Lite or Llama 4 Maverick?

Amazon Nova Lite is cheaper. It lists at $0.06 per million input tokens and $0.24 per million output tokens; Llama 4 Maverick lists at $0.19 and $0.65.

### Is Amazon Nova Lite or Llama 4 Maverick better for coding?

Amazon Nova Lite scores higher on coding benchmarks: 32.5 versus 26.6 in the Noometry coding category.

### Which has the bigger context window?

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

### How many benchmarks do Amazon Nova Lite and Llama 4 Maverick share?

26 benchmarks have published results for both models. Amazon Nova Lite has 34 scored results on Noometry and Llama 4 Maverick has 54.