Model comparison
Amazon Nova Lite vs GPT-4.1
GPT-4.1 is the stronger model overall, scoring 35.7 to 31.4 on the Noometry Index. Amazon Nova Lite costs 33× less per token, which makes it the better buy when GPT-4.1's lead doesn't matter for your workload.
Last verified . 26 shared benchmarks.
Summary
- They share 26 benchmarks with published results for both. Amazon Nova Lite scores higher in 2 categories and GPT-4.1 in 7 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where GPT-4.1 leads 57.8 to 42.3.
- The biggest single-benchmark swing is GPQA (HELM): 39.7% for Amazon Nova Lite and 65.9% for GPT-4.1.
- Amazon Nova Lite is cheaper at $0.06 / $0.24 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 Lite | GPT-4.1 | |
|---|---|---|
| Provider | Amazon | OpenAI |
| Noometry Index | 31.4 | 35.7 |
| Released | 2024-12-03 | 2025-04-14 |
| Weights | Proprietary | Proprietary |
| Context window | 300K | 1.05M |
| Max output | 10K | 33K |
| Input $ / M tokens | $0.06 | $2 |
| Output $ / M tokens | $0.24 | $8 |
| Results tracked | 34 | 52 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-4.1 leads
Amazon Nova Lite: 32.0 (#266), GPT-4.1: 34.2 (#228)
| Benchmark | Amazon Nova Lite | GPT-4.1 |
|---|---|---|
| LMArena Coding | 1239 | 1391 |
| ALE-Bench | 236.25 | 558.1 |
| SWE-bench Verified | — | 48.5% |
| SWE-bench Verified (bash only) | — | 39.6% |
| Aider Polyglot | — | 52.4% |
| WeirdML | — | 39% |
| LiveBench Coding | 27.5% | — |
| CadEval | — | 42% |
Agentic & Tool Use Not comparable
Amazon Nova Lite: —, GPT-4.1: 34.4 (#40)
| Benchmark | Amazon Nova Lite | GPT-4.1 |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 54% |
Reasoning Amazon Nova Lite leads
Amazon Nova Lite: 18.5 (#257), GPT-4.1: 11.7 (#334)
| Benchmark | Amazon Nova Lite | GPT-4.1 |
|---|---|---|
| LMArena Hard Prompts | 1220 | 1384 |
| 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 | 36.7% | — |
| DTBench | — | 68.3% |
| LiveBench Data Analysis | 37.2% | — |
| LMCA | — | 25.6% |
| Epoch Capabilities Index | — | 136.79 |
| ForecastBench | — | 61.5 |
| LiveBench | 36.4% | — |
Math Amazon Nova Lite leads
Amazon Nova Lite: 25.8 (#243), GPT-4.1: 21.4 (#269)
| Benchmark | Amazon Nova Lite | GPT-4.1 |
|---|---|---|
| Omni-MATH | 23.3% | 47.1% |
| LMArena Math | 1227 | 1369 |
| FrontierMath (Tiers 1-3) | — | 6% |
| OTIS Mock AIME 2024-2025 | — | 38.3% |
| LiveBench Math | 36.7% | — |
| MATH Level 5 | — | 83% |
| FrontierMath (Feb 2025 set) | — | 5.5% |
| FrontierMath Tier 4 (v1) | — | 0% |
Knowledge GPT-4.1 leads
Amazon Nova Lite: 26.0 (#254), GPT-4.1: 37.0 (#152)
| Benchmark | Amazon Nova Lite | GPT-4.1 |
|---|---|---|
| Humanity's Last Exam | 3.6% | 5.4% |
| MMLU-Pro | 60% | 81.1% |
| Vectara Hallucination Rate | 6.1% | 5.6% |
| GPQA (HELM) | 39.7% | 65.9% |
| LMArena Expert | 1201 | 1364 |
| GPQA Diamond | — | 66.9% |
| SimpleQA Verified | — | 31.1% |
| MMLU | 77% | — |
Multimodal GPT-4.1 leads
Amazon Nova Lite: 25.0 (#118), GPT-4.1: 37.5 (#66)
| Benchmark | Amazon Nova Lite | GPT-4.1 |
|---|---|---|
| LMArena Vision | 990 | 1210 |
| GeoBench | — | 72% |
Multilingual GPT-4.1 leads
Amazon Nova Lite: 38.2 (#227), GPT-4.1: 49.6 (#128)
| Benchmark | Amazon Nova Lite | GPT-4.1 |
|---|---|---|
| LMArena Non-English | 1208 | 1370 |
| LMArena Chinese | 1225 | 1382 |
| LMArena French | 1237 | 1383 |
| LMArena German | 1229 | 1380 |
| LMArena Japanese | 1153 | 1320 |
| LMArena Korean | 1153 | 1339 |
| LMArena Russian | 1216 | 1377 |
| LMArena Spanish | 1225 | 1375 |
Instruction Following GPT-4.1 leads
Amazon Nova Lite: 59.6 (#250), GPT-4.1: 71.5 (#148)
| Benchmark | Amazon Nova Lite | GPT-4.1 |
|---|---|---|
| IFEval | 77.6% | 83.8% |
| LMArena Instruction Following | 1205 | 1367 |
| LiveBench Instruction Following | 54.1% | — |
Long Context GPT-4.1 leads
Amazon Nova Lite: 37.4 (#212), GPT-4.1: 40.0 (#158)
| Benchmark | Amazon Nova Lite | GPT-4.1 |
|---|---|---|
| LMArena Longer Query | 1234 | 1385 |
| Fiction.LiveBench | — | 63.9% |
Writing & Preference GPT-4.1 leads
Amazon Nova Lite: 42.3 (#232), GPT-4.1: 57.8 (#120)
| Benchmark | Amazon Nova Lite | GPT-4.1 |
|---|---|---|
| LMArena Text | 1229 | 1383 |
| LMArena Creative Writing | 1197 | 1363 |
| WildBench | 75% | 85.4% |
| LMArena Multi-Turn | 1199 | 1398 |
| EQ-Bench Creative Writing | — | 1420 |
| LiveBench Language | 25.9% | — |
Frequently asked questions
Is Amazon Nova Lite better than GPT-4.1?
GPT-4.1 is the stronger model overall, scoring 35.7 to 31.4 on the Noometry Index. Amazon Nova Lite costs 33× 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 Lite or GPT-4.1?
Amazon Nova Lite is cheaper. It lists at $0.06 per million input tokens and $0.24 per million output tokens; GPT-4.1 lists at $2 and $8.
Is Amazon Nova Lite or GPT-4.1 better for coding?
GPT-4.1 scores higher on coding benchmarks: 34.2 versus 32.0 in the Noometry coding category.
Which has the bigger context window?
GPT-4.1 does, with 1.05M tokens against 300K.
How many benchmarks do Amazon Nova Lite and GPT-4.1 share?
26 benchmarks have published results for both models. Amazon Nova Lite has 34 scored results on Noometry and GPT-4.1 has 52.