Model comparison
Amazon Nova Pro vs GPT-4.1
GPT-4.1 is the stronger model overall, scoring 35.9 to 31.0 on the Noometry Index. Amazon Nova Pro costs 2.5× less per token, which makes it the better buy when GPT-4.1's lead doesn't matter for your workload.
Last verified . 27 shared benchmarks.
Summary
- They share 27 benchmarks with published results for both. Amazon Nova Pro scores higher in 3 categories and GPT-4.1 in 7 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in agentic & tool use, where GPT-4.1 leads 34.7 to 16.7.
- The biggest single-benchmark swing is Berkeley Function Calling Leaderboard: 25% for Amazon Nova Pro and 54% for GPT-4.1.
- Amazon Nova Pro is cheaper at $0.80 / $3.20 per million input/output tokens, against $2 / $8 for GPT-4.1.
- GPT-4.1 accepts more context: 1.05M tokens versus 300K.
Side by side
| Amazon Nova Pro | GPT-4.1 | |
|---|---|---|
| Provider | Amazon | OpenAI |
| Noometry Index | 31.0 | 35.9 |
| Released | 2024-12-03 | 2025-04-14 |
| Weights | Proprietary | Proprietary |
| Context window | 300K | 1.05M |
| Max output | 10K | 33K |
| Input $ / M tokens | $0.80 | $2 |
| Output $ / M tokens | $3.20 | $8 |
| Results tracked | 38 | 52 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Amazon Nova Pro: 35.1 (#229), GPT-4.1: 34.4 (#238)
| Benchmark | Amazon Nova Pro | GPT-4.1 |
|---|---|---|
| LMArena Coding | 1270 | 1391 |
| SWE-bench Verified | — | 48.5% |
| SWE-bench Verified (bash only) | — | 39.6% |
| Aider Polyglot | — | 52.4% |
| WeirdML | — | 39% |
| LiveBench Coding | 38.1% | — |
| CadEval | — | 42% |
| ALE-Bench | — | 558.1 |
Agentic & Tool Use GPT-4.1 leads
Amazon Nova Pro: 16.7 (#147), GPT-4.1: 34.7 (#43)
| Benchmark | Amazon Nova Pro | GPT-4.1 |
|---|---|---|
| Berkeley Function Calling Leaderboard | 25% | 54% |
| TheAgentCompany | 1.7% | — |
Reasoning Amazon Nova Pro leads
Amazon Nova Pro: 20.0 (#243), GPT-4.1: 11.7 (#339)
| Benchmark | Amazon Nova Pro | GPT-4.1 |
|---|---|---|
| LMArena Hard Prompts | 1246 | 1384 |
| Epoch Capabilities Index | 123.8 | 136.78 |
| ARC-AGI-2 | — | 0.4% |
| SimpleBench | — | 27% |
| Kagi LLM Benchmark | — | 52.3% |
| ARC-AGI-1 | — | 5.5% |
| Chess Puzzles | — | 6% |
| EnigmaEval | — | 2.2% |
| LiveBench Reasoning | 32.6% | — |
| DTBench | — | 68.3% |
| LiveBench Data Analysis | 48.3% | — |
| LMCA | — | 25.6% |
| ForecastBench | — | 61.5 |
| LiveBench | 43.5% | — |
Math Amazon Nova Pro leads
Amazon Nova Pro: 28.5 (#243), GPT-4.1: 22.3 (#280)
| Benchmark | Amazon Nova Pro | GPT-4.1 |
|---|---|---|
| Omni-MATH | 24.2% | 47.1% |
| LMArena Math | 1252 | 1370 |
| FrontierMath (Tiers 1-3) | — | 6% |
| OTIS Mock AIME 2024-2025 | — | 38.3% |
| LiveBench Math | 38% | — |
| MATH Level 5 | — | 83% |
| FrontierMath (Feb 2025 set) | — | 5.5% |
| FrontierMath Tier 4 (v1) | — | 0% |
Knowledge GPT-4.1 leads
Amazon Nova Pro: 27.4 (#250), GPT-4.1: 37.1 (#160)
| Benchmark | Amazon Nova Pro | GPT-4.1 |
|---|---|---|
| Humanity's Last Exam | 4.4% | 5.4% |
| MMLU-Pro | 67.3% | 81.1% |
| Vectara Hallucination Rate | 5.1% | 5.6% |
| GPQA (HELM) | 44.6% | 65.9% |
| LMArena Expert | 1211 | 1364 |
| GPQA Diamond | — | 66.9% |
| SimpleQA Verified | — | 31.1% |
| Confabulations | 30.1% | — |
| MMLU | 82% | — |
Multimodal GPT-4.1 leads
Amazon Nova Pro: 25.0 (#126), GPT-4.1: 38.2 (#67)
| Benchmark | Amazon Nova Pro | GPT-4.1 |
|---|---|---|
| LMArena Vision | 980 | 1211 |
| GeoBench | — | 72% |
Multilingual GPT-4.1 leads
Amazon Nova Pro: 39.7 (#223), GPT-4.1: 49.4 (#133)
| Benchmark | Amazon Nova Pro | GPT-4.1 |
|---|---|---|
| LMArena Non-English | 1234 | 1370 |
| LMArena Chinese | 1244 | 1382 |
| LMArena French | 1271 | 1382 |
| LMArena German | 1243 | 1381 |
| LMArena Japanese | 1200 | 1319 |
| LMArena Korean | 1203 | 1339 |
| LMArena Russian | 1240 | 1377 |
| LMArena Spanish | 1182 | 1376 |
Instruction Following GPT-4.1 leads
Amazon Nova Pro: 64.9 (#226), GPT-4.1: 71.3 (#153)
| Benchmark | Amazon Nova Pro | GPT-4.1 |
|---|---|---|
| IFEval | 81.5% | 83.8% |
| LMArena Instruction Following | 1235 | 1367 |
| LiveBench Instruction Following | 67.1% | — |
Long Context GPT-4.1 leads
Amazon Nova Pro: 38.1 (#205), GPT-4.1: 40.0 (#163)
| Benchmark | Amazon Nova Pro | GPT-4.1 |
|---|---|---|
| LMArena Longer Query | 1255 | 1385 |
| Fiction.LiveBench | — | 63.9% |
Writing & Preference GPT-4.1 leads
Amazon Nova Pro: 43.9 (#226), GPT-4.1: 57.6 (#125)
| Benchmark | Amazon Nova Pro | GPT-4.1 |
|---|---|---|
| LMArena Text | 1259 | 1383 |
| LMArena Creative Writing | 1212 | 1363 |
| WildBench | 77.7% | 85.4% |
| LMArena Multi-Turn | 1246 | 1398 |
| Short-Story Creative Writing | 60.5% | — |
| EQ-Bench Creative Writing | — | 1420 |
| LiveBench Language | 37% | — |
Frequently asked questions
Is Amazon Nova Pro better than GPT-4.1?
GPT-4.1 is the stronger model overall, scoring 35.9 to 31.0 on the Noometry Index. Amazon Nova Pro costs 2.5× less per token, which makes it the better buy when GPT-4.1's lead doesn't matter for your workload.
Which is cheaper, Amazon Nova Pro or GPT-4.1?
Amazon Nova Pro is cheaper. It lists at $0.80 per million input tokens and $3.20 per million output tokens; GPT-4.1 lists at $2 and $8.
Is Amazon Nova Pro or GPT-4.1 better for coding?
They score almost the same on coding (35.1 vs 34.4); test both on your own repository before choosing.
Which has the bigger context window?
GPT-4.1 does, with 1.05M tokens against 300K.
How many benchmarks do Amazon Nova Pro and GPT-4.1 share?
27 benchmarks have published results for both models. Amazon Nova Pro has 38 scored results on Noometry and GPT-4.1 has 52.