Model comparison
Amazon Nova Lite vs GPT-4o
Amazon Nova Lite is the stronger model overall, scoring 31.9 to 28.6 on the Noometry Index.
Last verified . 33 shared benchmarks.
Summary
- They share 33 benchmarks with published results for both. Amazon Nova Lite scores higher in 3 categories and GPT-4o in 6 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Amazon Nova Lite leads 27.8 to 10.6.
- The biggest single-benchmark swing is LiveBench Coding: 27.5% for Amazon Nova Lite and 51.4% for GPT-4o.
- Amazon Nova Lite is cheaper at $0.06 / $0.24 per million input/output tokens, against $2.50 / $10 for GPT-4o.
- Amazon Nova Lite accepts more context: 300K tokens versus 128K.
Side by side
| Amazon Nova Lite | GPT-4o | |
|---|---|---|
| Provider | Amazon | OpenAI |
| Noometry Index | 31.9 | 28.6 |
| Released | 2024-12-03 | 2024-05-13 |
| Weights | Proprietary | Proprietary |
| Context window | 300K | 128K |
| Max output | 10K | 16K |
| Input $ / M tokens | $0.06 | $2.50 |
| Output $ / M tokens | $0.24 | $10 |
| Results tracked | 34 | 72 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Amazon Nova Lite leads
Amazon Nova Lite: 32.5 (#270), GPT-4o: 24.8 (#328)
| Benchmark | Amazon Nova Lite | GPT-4o |
|---|---|---|
| LiveBench Coding | 27.5% | 51.4% |
| LMArena Coding | 1239 | 1297 |
| SWE-bench Verified | — | 31% |
| SWE-bench Verified (bash only) | — | 21.6% |
| Aider Polyglot | — | 45.3% |
| GSO | — | 0% |
| WeirdML | — | 25.1% |
| BigCodeBench Instruct | — | 51.1% |
| BigCodeBench Complete | — | 61.1% |
| CadEval | — | 26% |
| ALE-Bench | 236.25 | — |
| HumanEval+ | — | 87.2% |
| MBPP+ | — | 72.2% |
Agentic & Tool Use Not comparable
Amazon Nova Lite: —, GPT-4o: 21.0 (#141)
| Benchmark | Amazon Nova Lite | GPT-4o |
|---|---|---|
| GDPval | — | 9.9% |
| TheAgentCompany | — | 8.6% |
| Cybench | — | 12.5% |
| BALROG | — | 32.3% |
| LMArena Search | — | 1006 |
| METR Time Horizons | — | 40.8% |
Reasoning Amazon Nova Lite leads
Amazon Nova Lite: 19.3 (#260), GPT-4o: 9.4 (#343)
| Benchmark | Amazon Nova Lite | GPT-4o |
|---|---|---|
| LiveBench Reasoning | 36.7% | 55.8% |
| LMArena Hard Prompts | 1220 | 1281 |
| LiveBench Data Analysis | 37.2% | 60.9% |
| LiveBench | 36.4% | 55.3% |
| ARC-AGI-2 | — | 0% |
| SimpleBench | — | 17.8% |
| ARC-AGI-1 | — | 4.5% |
| CritPt | — | 0% |
| Chess Puzzles | — | 13% |
| EnigmaEval | — | 0.8% |
| DTBench | — | 64.5% |
| LMCA | — | 16.6% |
| Epoch Capabilities Index | — | 128.97 |
| ForecastBench | — | 57.7 |
Math Amazon Nova Lite leads
Amazon Nova Lite: 27.8 (#247), GPT-4o: 10.6 (#312)
| Benchmark | Amazon Nova Lite | GPT-4o |
|---|---|---|
| Omni-MATH | 23.3% | 29.3% |
| LiveBench Math | 36.7% | 49.5% |
| LMArena Math | 1227 | 1285 |
| FrontierMath (Tiers 1-3) | — | 0.4% |
| OTIS Mock AIME 2024-2025 | — | 6.4% |
| MATH Level 5 | — | 53.3% |
| FrontierMath (Feb 2025 set) | — | 0.3% |
Knowledge GPT-4o leads
Amazon Nova Lite: 26.3 (#259), GPT-4o: 28.8 (#242)
| Benchmark | Amazon Nova Lite | GPT-4o |
|---|---|---|
| Humanity's Last Exam | 3.6% | 2.7% |
| MMLU-Pro | 60% | 71.3% |
| Vectara Hallucination Rate | 6.1% | 9.6% |
| GPQA (HELM) | 39.7% | 52% |
| LMArena Expert | 1201 | 1250 |
| MMLU | 77% | 88.1% |
| GPQA Diamond | — | 49.2% |
| SimpleQA Verified | — | 26% |
| Confabulations | — | 15.3% |
Multimodal GPT-4o leads
Amazon Nova Lite: 25.5 (#123), GPT-4o: 34.5 (#91)
| Benchmark | Amazon Nova Lite | GPT-4o |
|---|---|---|
| LMArena Vision | 990 | 1137 |
| Video-MME | — | 71.9% |
| GeoBench | — | 71% |
| VPCT | — | 40% |
| ScienceQA | — | 88.5% |
Multilingual GPT-4o leads
Amazon Nova Lite: 38.0 (#232), GPT-4o: 43.2 (#186)
| Benchmark | Amazon Nova Lite | GPT-4o |
|---|---|---|
| LMArena Non-English | 1208 | 1283 |
| LMArena Chinese | 1225 | 1277 |
| LMArena French | 1237 | 1304 |
| LMArena German | 1229 | 1282 |
| LMArena Japanese | 1153 | 1257 |
| LMArena Korean | 1153 | 1234 |
| LMArena Russian | 1216 | 1286 |
| LMArena Spanish | 1226 | 1292 |
Instruction Following GPT-4o leads
Amazon Nova Lite: 59.3 (#255), GPT-4o: 66.6 (#207)
| Benchmark | Amazon Nova Lite | GPT-4o |
|---|---|---|
| LiveBench Instruction Following | 54.1% | 68.6% |
| IFEval | 77.6% | 81.7% |
| LMArena Instruction Following | 1205 | 1278 |
Long Context GPT-4o leads
Amazon Nova Lite: 37.4 (#217), GPT-4o: 39.4 (#179)
| Benchmark | Amazon Nova Lite | GPT-4o |
|---|---|---|
| LMArena Longer Query | 1234 | 1289 |
| Fiction.LiveBench | — | 66.7% |
Writing & Preference GPT-4o leads
Amazon Nova Lite: 42.1 (#237), GPT-4o: 52.6 (#166)
| Benchmark | Amazon Nova Lite | GPT-4o |
|---|---|---|
| LMArena Text | 1229 | 1300 |
| LMArena Creative Writing | 1197 | 1292 |
| WildBench | 75% | 82.8% |
| LMArena Multi-Turn | 1199 | 1302 |
| LiveBench Language | 25.9% | 47.6% |
| Short-Story Creative Writing | — | 81.8% |
Frequently asked questions
Is Amazon Nova Lite better than GPT-4o?
Amazon Nova Lite is the stronger model overall, scoring 31.9 to 28.6 on the Noometry Index.
Which is cheaper, Amazon Nova Lite or GPT-4o?
Amazon Nova Lite is cheaper. It lists at $0.06 per million input tokens and $0.24 per million output tokens; GPT-4o lists at $2.50 and $10.
Is Amazon Nova Lite or GPT-4o better for coding?
Amazon Nova Lite scores higher on coding benchmarks: 32.5 versus 24.8 in the Noometry coding category.
Which has the bigger context window?
Amazon Nova Lite does, with 300K tokens against 128K.
How many benchmarks do Amazon Nova Lite and GPT-4o share?
33 benchmarks have published results for both models. Amazon Nova Lite has 34 scored results on Noometry and GPT-4o has 72.