# Amazon Nova Lite vs Qwen3 235B-A22B

> Qwen3 235B-A22B is the stronger model overall, scoring 43.5 to 31.9 on the Noometry Index. Amazon Nova Lite costs 12× less per token, which makes it the better buy when Qwen3 235B-A22B's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/amazon-nova-lite-vs-qwen3-235b-a22b
- Last updated: 2026-10-10
- Shared benchmarks: 23

## Summary

- They share 23 benchmarks with published results for both. Amazon Nova Lite scores higher in 1 category and Qwen3 235B-A22B in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen3 235B-A22B leads 49.6 to 26.3.
- The biggest single-benchmark swing is Omni-MATH: 23.3% for Amazon Nova Lite and 71.8% for Qwen3 235B-A22B.
- Amazon Nova Lite is cheaper at $0.06 / $0.24 per million input/output tokens, against $0.70 / $2.80 for Qwen3 235B-A22B.
- Amazon Nova Lite accepts more context: 300K tokens versus 131K.
- Qwen3 235B-A22B has downloadable open weights; the other is API-only.

## Snapshot

| | Amazon Nova Lite | Qwen3 235B-A22B |
|---|---|---|
| Provider | Amazon | Alibaba (Qwen) |
| Noometry Index | 31.9 | 43.5 |
| Rank | 265 | 91 |
| Context | 300K | 131K |
| Input $/M | $0.06 | $0.70 |
| Output $/M | $0.24 | $2.80 |
| Weights | Proprietary | Open |

## Coding

- Amazon Nova Lite: 32.5 (#270)
- Qwen3 235B-A22B: 44.3 (#75)

| Benchmark | Amazon Nova Lite | Qwen3 235B-A22B |
|---|---|---|
| LMArena Coding | 1239 | 1445 |
| Aider Polyglot | — | 59.6% |
| SciCode | — | 42.4% |
| WeirdML | — | 41% |
| LiveBench Coding | 27.5% | — |
| ALE-Bench | 236.25 | — |

## Agentic & Tool Use

- Amazon Nova Lite: —
- Qwen3 235B-A22B: 33.9 (#51)

| Benchmark | Amazon Nova Lite | Qwen3 235B-A22B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 52.1% |
| Vending-Bench 2 | — | -11.34 |

## Reasoning

- Amazon Nova Lite: 19.3 (#260)
- Qwen3 235B-A22B: 15.7 (#311)

| Benchmark | Amazon Nova Lite | Qwen3 235B-A22B |
|---|---|---|
| LMArena Hard Prompts | 1220 | 1433 |
| ARC-AGI-2 | — | 1.3% |
| SimpleBench | — | 31% |
| Kagi LLM Benchmark | — | 69.4% |
| ARC-AGI-1 | — | 11% |
| CritPt | — | 0% |
| Chess Puzzles | — | 12% |
| LiveBench Reasoning | 36.7% | — |
| Mystery Game Puzzles | — | 9% |
| DTBench | — | 80.3% |
| LiveBench Data Analysis | 37.2% | — |
| LMCA | — | 29.3% |
| Epoch Capabilities Index | — | 143.85 |
| ForecastBench | — | 59.7 |
| LiveBench | 36.4% | — |

## Math

- Amazon Nova Lite: 27.8 (#247)
- Qwen3 235B-A22B: 50.4 (#57)

| Benchmark | Amazon Nova Lite | Qwen3 235B-A22B |
|---|---|---|
| Omni-MATH | 23.3% | 71.8% |
| LMArena Math | 1227 | 1432 |
| OTIS Mock AIME 2024-2025 | — | 86.7% |
| LiveBench Math | 36.7% | — |
| MATH Level 5 | — | 68.9% |
| FrontierMath (Feb 2025 set) | — | 8.5% |
| FrontierMath Tier 4 (v1) | — | 0% |

## Knowledge

- Amazon Nova Lite: 26.3 (#259)
- Qwen3 235B-A22B: 49.6 (#73)

| Benchmark | Amazon Nova Lite | Qwen3 235B-A22B |
|---|---|---|
| MMLU-Pro | 60% | 84.4% |
| Vectara Hallucination Rate | 6.1% | 9.3% |
| GPQA (HELM) | 39.7% | 72.7% |
| LMArena Expert | 1201 | 1463 |
| GPQA Diamond | — | 80.1% |
| Humanity's Last Exam | 3.6% | — |
| SimpleQA Verified | — | 40.4% |
| Confabulations | — | 15.6% |
| MMLU | 77% | — |

## Multimodal

- Amazon Nova Lite: 25.5 (#123)
- Qwen3 235B-A22B: —

| Benchmark | Amazon Nova Lite | Qwen3 235B-A22B |
|---|---|---|
| LMArena Vision | 990 | — |

## Multilingual

- Amazon Nova Lite: 38.0 (#232)
- Qwen3 235B-A22B: 52.3 (#89)

| Benchmark | Amazon Nova Lite | Qwen3 235B-A22B |
|---|---|---|
| LMArena Non-English | 1208 | 1409 |
| LMArena Chinese | 1225 | 1481 |
| LMArena French | 1237 | 1445 |
| LMArena German | 1229 | 1433 |
| LMArena Japanese | 1153 | 1399 |
| LMArena Korean | 1153 | 1391 |
| LMArena Russian | 1216 | 1411 |
| LMArena Spanish | 1226 | 1430 |

## Instruction Following

- Amazon Nova Lite: 59.3 (#255)
- Qwen3 235B-A22B: 72.6 (#136)

| Benchmark | Amazon Nova Lite | Qwen3 235B-A22B |
|---|---|---|
| IFEval | 77.6% | 83.5% |
| LMArena Instruction Following | 1205 | 1408 |
| LiveBench Instruction Following | 54.1% | — |

## Long Context

- Amazon Nova Lite: 37.4 (#217)
- Qwen3 235B-A22B: 46.1 (#26)

| Benchmark | Amazon Nova Lite | Qwen3 235B-A22B |
|---|---|---|
| LMArena Longer Query | 1234 | 1426 |
| Fiction.LiveBench | — | 75% |

## Writing & Preference

- Amazon Nova Lite: 42.1 (#237)
- Qwen3 235B-A22B: 59.6 (#108)

| Benchmark | Amazon Nova Lite | Qwen3 235B-A22B |
|---|---|---|
| LMArena Text | 1229 | 1419 |
| LMArena Creative Writing | 1197 | 1384 |
| WildBench | 75% | 86.6% |
| LMArena Multi-Turn | 1199 | 1432 |
| Short-Story Creative Writing | — | 83% |
| EQ-Bench Creative Writing | — | 1366 |
| LiveBench Language | 25.9% | — |

## FAQ

### Is Amazon Nova Lite better than Qwen3 235B-A22B?

Qwen3 235B-A22B is the stronger model overall, scoring 43.5 to 31.9 on the Noometry Index. Amazon Nova Lite costs 12× less per token, which makes it the better buy when Qwen3 235B-A22B's lead doesn't matter for your workload.

### Which is cheaper, Amazon Nova Lite or Qwen3 235B-A22B?

Amazon Nova Lite is cheaper. It lists at $0.06 per million input tokens and $0.24 per million output tokens; Qwen3 235B-A22B lists at $0.70 and $2.80.

### Is Amazon Nova Lite or Qwen3 235B-A22B better for coding?

Qwen3 235B-A22B scores higher on coding benchmarks: 44.3 versus 32.5 in the Noometry coding category.

### Which has the bigger context window?

Amazon Nova Lite does, with 300K tokens against 131K.

### How many benchmarks do Amazon Nova Lite and Qwen3 235B-A22B share?

23 benchmarks have published results for both models. Amazon Nova Lite has 34 scored results on Noometry and Qwen3 235B-A22B has 49.