Model comparison
Amazon Nova Pro vs Claude Sonnet 4
Claude Sonnet 4 is the stronger model overall, scoring 40.8 to 31.0 on the Noometry Index. Amazon Nova Pro costs 4.3× less per token, which makes it the better buy when Claude Sonnet 4's lead doesn't matter for your workload.
Last verified . 29 shared benchmarks.
Summary
- They share 29 benchmarks with published results for both. Amazon Nova Pro scores higher in 1 category and Claude Sonnet 4 in 9 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in agentic & tool use, where Claude Sonnet 4 leads 38.5 to 16.7.
- The biggest single-benchmark swing is Omni-MATH: 24.2% for Amazon Nova Pro and 60.2% for Claude Sonnet 4.
- Amazon Nova Pro is cheaper at $0.80 / $3.20 per million input/output tokens, against $3 / $15 for Claude Sonnet 4.
- Amazon Nova Pro accepts more context: 300K tokens versus 200K.
Side by side
| Amazon Nova Pro | Claude Sonnet 4 | |
|---|---|---|
| Provider | Amazon | Anthropic |
| Noometry Index | 31.0 | 40.8 |
| Released | 2024-12-03 | 2025-05-22 |
| Weights | Proprietary | Proprietary |
| Context window | 300K | 200K |
| Max output | 10K | 64K |
| Input $ / M tokens | $0.80 | $3 |
| Output $ / M tokens | $3.20 | $15 |
| Results tracked | 38 | 58 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Sonnet 4 leads
Amazon Nova Pro: 35.1 (#229), Claude Sonnet 4: 43.5 (#88)
| Benchmark | Amazon Nova Pro | Claude Sonnet 4 |
|---|---|---|
| LMArena Coding | 1270 | 1414 |
| SWE-bench Verified (bash only) | — | 64.9% |
| Aider Polyglot | — | 61.3% |
| SciCode | — | 40% |
| GSO | — | 4.9% |
| WeirdML | — | 46.1% |
| LiveBench Coding | 38.1% | — |
| ALE-Bench | — | 655.35 |
Agentic & Tool Use Claude Sonnet 4 leads
Amazon Nova Pro: 16.7 (#147), Claude Sonnet 4: 38.5 (#31)
| Benchmark | Amazon Nova Pro | Claude Sonnet 4 |
|---|---|---|
| TheAgentCompany | 1.7% | 33.1% |
| Berkeley Function Calling Leaderboard | 25% | — |
| Cybench | — | 35% |
| DeepResearch Bench | — | 46.6% |
| OSWorld | — | 43.9% |
| METR Time Horizons | — | 62% |
Reasoning Claude Sonnet 4 leads
Amazon Nova Pro: 20.0 (#243), Claude Sonnet 4: 22.9 (#187)
| Benchmark | Amazon Nova Pro | Claude Sonnet 4 |
|---|---|---|
| LMArena Hard Prompts | 1246 | 1372 |
| Epoch Capabilities Index | 123.8 | 141.69 |
| ARC-AGI-2 | — | 5.9% |
| SimpleBench | — | 45.5% |
| Kagi LLM Benchmark | — | 73% |
| ARC-AGI-1 | — | 40% |
| CritPt | — | 0.3% |
| EnigmaEval | — | 3.1% |
| LiveBench Reasoning | 32.6% | — |
| DTBench | — | 77.1% |
| LiveBench Data Analysis | 48.3% | — |
| LMCA | — | 29% |
| ForecastBench | — | 60.2 |
| LiveBench | 43.5% | — |
Math Claude Sonnet 4 leads
Amazon Nova Pro: 28.5 (#243), Claude Sonnet 4: 43.3 (#80)
| Benchmark | Amazon Nova Pro | Claude Sonnet 4 |
|---|---|---|
| Omni-MATH | 24.2% | 60.2% |
| LMArena Math | 1252 | 1375 |
| OTIS Mock AIME 2024-2025 | — | 71.1% |
| LiveBench Math | 38% | — |
| MATH Level 5 | — | 84.4% |
| FrontierMath (Feb 2025 set) | — | 4.1% |
| FrontierMath Tier 4 (v1) | — | 0% |
Knowledge Claude Sonnet 4 leads
Amazon Nova Pro: 27.4 (#250), Claude Sonnet 4: 41.8 (#108)
| Benchmark | Amazon Nova Pro | Claude Sonnet 4 |
|---|---|---|
| Humanity's Last Exam | 4.4% | 7.8% |
| MMLU-Pro | 67.3% | 84.3% |
| Confabulations | 30.1% | 13.2% |
| Vectara Hallucination Rate | 5.1% | 10.3% |
| GPQA (HELM) | 44.6% | 70.6% |
| LMArena Expert | 1211 | 1372 |
| GPQA Diamond | — | 79.2% |
| MMLU | 82% | — |
Multimodal Claude Sonnet 4 leads
Amazon Nova Pro: 25.0 (#126), Claude Sonnet 4: 26.2 (#121)
| Benchmark | Amazon Nova Pro | Claude Sonnet 4 |
|---|---|---|
| LMArena Vision | 980 | 1191 |
| GeoBench | — | 37% |
| VPCT | — | 34% |
| MindCube | — | 44.8% |
Multilingual Claude Sonnet 4 leads
Amazon Nova Pro: 39.7 (#223), Claude Sonnet 4: 46.7 (#156)
| Benchmark | Amazon Nova Pro | Claude Sonnet 4 |
|---|---|---|
| LMArena Non-English | 1234 | 1333 |
| LMArena Chinese | 1244 | 1350 |
| LMArena French | 1271 | 1363 |
| LMArena German | 1243 | 1331 |
| LMArena Japanese | 1200 | 1302 |
| LMArena Korean | 1203 | 1291 |
| LMArena Russian | 1240 | 1355 |
| LMArena Spanish | 1182 | 1357 |
Instruction Following Claude Sonnet 4 leads
Amazon Nova Pro: 64.9 (#226), Claude Sonnet 4: 71.7 (#145)
| Benchmark | Amazon Nova Pro | Claude Sonnet 4 |
|---|---|---|
| IFEval | 81.5% | 84% |
| LMArena Instruction Following | 1235 | 1376 |
| LiveBench Instruction Following | 67.1% | — |
Long Context Amazon Nova Pro leads
Amazon Nova Pro: 38.1 (#205), Claude Sonnet 4: 33.7 (#259)
| Benchmark | Amazon Nova Pro | Claude Sonnet 4 |
|---|---|---|
| LMArena Longer Query | 1255 | 1398 |
| Fiction.LiveBench | — | 46.9% |
Writing & Preference Claude Sonnet 4 leads
Amazon Nova Pro: 43.9 (#226), Claude Sonnet 4: 57.1 (#132)
| Benchmark | Amazon Nova Pro | Claude Sonnet 4 |
|---|---|---|
| LMArena Text | 1259 | 1351 |
| LMArena Creative Writing | 1212 | 1345 |
| Short-Story Creative Writing | 60.5% | 81.4% |
| WildBench | 77.7% | 83.8% |
| LMArena Multi-Turn | 1246 | 1376 |
| EQ-Bench Creative Writing | — | 1483 |
| LiveBench Language | 37% | — |
Frequently asked questions
Is Amazon Nova Pro better than Claude Sonnet 4?
Claude Sonnet 4 is the stronger model overall, scoring 40.8 to 31.0 on the Noometry Index. Amazon Nova Pro costs 4.3× less per token, which makes it the better buy when Claude Sonnet 4's lead doesn't matter for your workload.
Which is cheaper, Amazon Nova Pro or Claude Sonnet 4?
Amazon Nova Pro is cheaper. It lists at $0.80 per million input tokens and $3.20 per million output tokens; Claude Sonnet 4 lists at $3 and $15.
Is Amazon Nova Pro or Claude Sonnet 4 better for coding?
Claude Sonnet 4 scores higher on coding benchmarks: 43.5 versus 35.1 in the Noometry coding category.
Which has the bigger context window?
Amazon Nova Pro does, with 300K tokens against 200K.
How many benchmarks do Amazon Nova Pro and Claude Sonnet 4 share?
29 benchmarks have published results for both models. Amazon Nova Pro has 38 scored results on Noometry and Claude Sonnet 4 has 58.