# Amazon Nova Micro vs Claude Sonnet 4.6

> Claude Sonnet 4.6 is the stronger model overall, scoring 50.3 to 30.4 on the Noometry Index. Amazon Nova Micro costs 98× less per token, which makes it the better buy when Claude Sonnet 4.6's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/amazon-nova-micro-vs-claude-sonnet-4-6
- Last updated: 2026-10-10
- Shared benchmarks: 18

## Summary

- They share 18 benchmarks with published results for both. Amazon Nova Micro scores higher in 0 categories and Claude Sonnet 4.6 in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Claude Sonnet 4.6 leads 70.2 to 39.5.
- The biggest single-benchmark swing is Vectara Hallucination Rate: 5.5% for Amazon Nova Micro and 10.6% for Claude Sonnet 4.6.
- Amazon Nova Micro is cheaper at $0.035 / $0.14 per million input/output tokens, against $3 / $15 for Claude Sonnet 4.6.
- Claude Sonnet 4.6 accepts more context: 1M tokens versus 128K.

## Snapshot

| | Amazon Nova Micro | Claude Sonnet 4.6 |
|---|---|---|
| Provider | Amazon | Anthropic |
| Noometry Index | 30.4 | 50.3 |
| Rank | 294 | 50 |
| Context | 128K | 1M |
| Input $/M | $0.035 | $3 |
| Output $/M | $0.14 | $15 |
| Weights | Proprietary | Proprietary |

## Coding

- Amazon Nova Micro: 30.5 (#295)
- Claude Sonnet 4.6: 46.3 (#67)

| Benchmark | Amazon Nova Micro | Claude Sonnet 4.6 |
|---|---|---|
| LMArena Coding | 1218 | 1504 |
| SWE-bench Verified | — | 75.2% |
| DeepSWE | — | 29.9% |
| FrontierCode | — | 24.3% |
| LMArena WebDev | — | 1522 |
| SciCode | — | 46.8% |
| WeirdML | — | 66.1% |
| LiveBench Coding | 20.2% | — |
| ALE-Bench | — | 1,327 |

## Agentic & Tool Use

- Amazon Nova Micro: 22.1 (#132)
- Claude Sonnet 4.6: 39.1 (#28)

| Benchmark | Amazon Nova Micro | Claude Sonnet 4.6 |
|---|---|---|
| Terminal-Bench | — | 53.4% |
| APEX-Agents | — | 43% |
| Berkeley Function Calling Leaderboard | 22.3% | — |
| OSWorld 2.0 | — | 9.3% |
| DeepResearch Bench | — | 54.9% |
| OSWorld | — | 72.1% |
| ExploitBench | — | 23.6% |
| GBAEval | — | 48.8% |
| GDP.pdf | — | 18% |
| LMArena Search | — | 1221 |
| Vending-Bench 2 | — | 7,204 |

## Reasoning

- Amazon Nova Micro: 17.4 (#294)
- Claude Sonnet 4.6: 46.1 (#45)

| Benchmark | Amazon Nova Micro | Claude Sonnet 4.6 |
|---|---|---|
| LMArena Hard Prompts | 1191 | 1484 |
| ARC-AGI-2 | — | 60.4% |
| NYT Connections (extended) | — | 80.9% |
| ARC-AGI-1 | — | 86.5% |
| CritPt | — | 3.1% |
| Chess Puzzles | — | 13% |
| Thematic Generalization | — | 76.3% |
| LiveBench Reasoning | 25.1% | — |
| Mystery Game Puzzles | — | 16% |
| DTBench | — | 89.9% |
| LiveBench Data Analysis | 34% | — |
| LMCA | — | 46.5% |
| Epoch Capabilities Index | — | 152.24 |
| ForecastBench | — | 62 |
| LiveBench | 29.6% | — |

## Math

- Amazon Nova Micro: 26.9 (#254)
- Claude Sonnet 4.6: 52.9 (#49)

| Benchmark | Amazon Nova Micro | Claude Sonnet 4.6 |
|---|---|---|
| LMArena Math | 1206 | 1462 |
| OTIS Mock AIME 2024-2025 | — | 85.8% |
| ProofBench | — | 45% |
| Omni-MATH | 21.4% | — |
| LiveBench Math | 34.5% | — |
| FrontierMath (Feb 2025 set) | — | 32.4% |
| FrontierMath Tier 4 (v1) | — | 8.3% |

## Knowledge

- Amazon Nova Micro: 29.6 (#237)
- Claude Sonnet 4.6: 51.7 (#65)

| Benchmark | Amazon Nova Micro | Claude Sonnet 4.6 |
|---|---|---|
| Vectara Hallucination Rate | 5.5% | 10.6% |
| LMArena Expert | 1184 | 1500 |
| GPQA Diamond | — | 87.4% |
| SimpleQA Verified | — | 35.5% |
| MMLU-Pro | 51.1% | — |
| GPQA (HELM) | 38.3% | — |
| MMLU | 70.8% | — |

## Multimodal

- Amazon Nova Micro: —
- Claude Sonnet 4.6: 38.0 (#68)

| Benchmark | Amazon Nova Micro | Claude Sonnet 4.6 |
|---|---|---|
| LMArena Vision | — | 1283 |
| Blueprint-Bench 2 | — | 6.7% |
| LMArena Document | — | 1482 |

## Multilingual

- Amazon Nova Micro: 36.5 (#239)
- Claude Sonnet 4.6: 54.4 (#41)

| Benchmark | Amazon Nova Micro | Claude Sonnet 4.6 |
|---|---|---|
| LMArena Non-English | 1186 | 1440 |
| LMArena Chinese | 1209 | 1491 |
| LMArena French | 1238 | 1465 |
| LMArena German | 1192 | 1428 |
| LMArena Japanese | 1154 | 1420 |
| LMArena Korean | 1150 | 1411 |
| LMArena Russian | 1185 | 1440 |
| LMArena Spanish | 1225 | 1464 |

## Instruction Following

- Amazon Nova Micro: 56.3 (#272)
- Claude Sonnet 4.6: 77.4 (#25)

| Benchmark | Amazon Nova Micro | Claude Sonnet 4.6 |
|---|---|---|
| LMArena Instruction Following | 1174 | 1475 |
| LiveBench Instruction Following | 48% | — |
| IFEval | 76% | — |

## Long Context

- Amazon Nova Micro: 36.5 (#229)
- Claude Sonnet 4.6: 45.3 (#44)

| Benchmark | Amazon Nova Micro | Claude Sonnet 4.6 |
|---|---|---|
| LMArena Longer Query | 1205 | 1479 |

## Writing & Preference

- Amazon Nova Micro: 39.5 (#247)
- Claude Sonnet 4.6: 70.2 (#22)

| Benchmark | Amazon Nova Micro | Claude Sonnet 4.6 |
|---|---|---|
| LMArena Text | 1208 | 1458 |
| LMArena Creative Writing | 1172 | 1435 |
| LMArena Multi-Turn | 1178 | 1464 |
| EQ-Bench Creative Writing | — | 1810 |
| WildBench | 74.3% | — |
| EQ-Bench 4 | — | 1207 |
| LiveBench Language | 15.8% | — |

## FAQ

### Is Amazon Nova Micro better than Claude Sonnet 4.6?

Claude Sonnet 4.6 is the stronger model overall, scoring 50.3 to 30.4 on the Noometry Index. Amazon Nova Micro costs 98× less per token, which makes it the better buy when Claude Sonnet 4.6's lead doesn't matter for your workload.

### Which is cheaper, Amazon Nova Micro or Claude Sonnet 4.6?

Amazon Nova Micro is cheaper. It lists at $0.035 per million input tokens and $0.14 per million output tokens; Claude Sonnet 4.6 lists at $3 and $15.

### Is Amazon Nova Micro or Claude Sonnet 4.6 better for coding?

Claude Sonnet 4.6 scores higher on coding benchmarks: 46.3 versus 30.5 in the Noometry coding category.

### Which has the bigger context window?

Claude Sonnet 4.6 does, with 1M tokens against 128K.

### How many benchmarks do Amazon Nova Micro and Claude Sonnet 4.6 share?

18 benchmarks have published results for both models. Amazon Nova Micro has 32 scored results on Noometry and Claude Sonnet 4.6 has 57.