Model comparison
Amazon Nova Lite vs GPT-3.5-turbo
Amazon Nova Lite is the stronger model overall, scoring 31.4 to 23.1 on the Noometry Index.
Last verified . 18 shared benchmarks.
Summary
- They share 18 benchmarks with published results for both. Amazon Nova Lite scores higher in 8 categories and GPT-3.5-turbo in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Amazon Nova Lite leads 25.8 to 6.0.
- Amazon Nova Lite is cheaper at $0.06 / $0.24 per million input/output tokens, against $0.50 / $1.50 for GPT-3.5-turbo.
- Amazon Nova Lite accepts more context: 300K tokens versus 16K.
Side by side
| Amazon Nova Lite | GPT-3.5-turbo | |
|---|---|---|
| Provider | Amazon | OpenAI |
| Noometry Index | 31.4 | 23.1 |
| Released | 2024-12-03 | 2023-03-01 |
| Weights | Proprietary | Proprietary |
| Context window | 300K | 16K |
| Max output | 10K | 4K |
| Input $ / M tokens | $0.06 | $0.50 |
| Output $ / M tokens | $0.24 | $1.50 |
| Results tracked | 34 | 44 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Amazon Nova Lite leads
Amazon Nova Lite: 32.0 (#266), GPT-3.5-turbo: 23.5 (#326)
| Benchmark | Amazon Nova Lite | GPT-3.5-turbo |
|---|---|---|
| LMArena Coding | 1239 | 1137 |
| WeirdML | — | 3.5% |
| BigCodeBench Instruct | — | 39.1% |
| LiveBench Coding | 27.5% | — |
| BigCodeBench Complete | — | 50.6% |
| ALE-Bench | 236.25 | — |
| HumanEval+ | — | 70.7% |
| MBPP+ | — | 69.7% |
Agentic & Tool Use Not comparable
Amazon Nova Lite: —, GPT-3.5-turbo: —
| Benchmark | Amazon Nova Lite | GPT-3.5-turbo |
|---|---|---|
| METR Time Horizons | — | 21.5% |
Reasoning Amazon Nova Lite leads
Amazon Nova Lite: 18.5 (#257), GPT-3.5-turbo: 13.3 (#328)
| Benchmark | Amazon Nova Lite | GPT-3.5-turbo |
|---|---|---|
| LMArena Hard Prompts | 1220 | 1108 |
| Chess Puzzles | — | 0% |
| LiveBench Reasoning | 36.7% | — |
| Mystery Game Puzzles | — | 3% |
| DTBench | — | 48.5% |
| LiveBench Data Analysis | 37.2% | — |
| LMCA | — | 9.7% |
| Adversarial NLI | — | 58.1% |
| BIG-Bench Hard | — | 61.6% |
| CommonsenseQA 2.0 | — | 57% |
| Epoch Capabilities Index | — | 118.48 |
| ForecastBench | — | 50.4 |
| LiveBench | 36.4% | — |
| WinoGrande | — | 81.6% |
Math Amazon Nova Lite leads
Amazon Nova Lite: 25.8 (#243), GPT-3.5-turbo: 6.0 (#321)
| Benchmark | Amazon Nova Lite | GPT-3.5-turbo |
|---|---|---|
| LMArena Math | 1227 | 1142 |
| FrontierMath (Tiers 1-3) | — | 0% |
| OTIS Mock AIME 2024-2025 | — | 2.2% |
| Omni-MATH | 23.3% | — |
| LiveBench Math | 36.7% | — |
| MATH Level 5 | — | 15.9% |
| GSM8K | — | 57.8% |
Knowledge Amazon Nova Lite leads
Amazon Nova Lite: 26.0 (#254), GPT-3.5-turbo: 10.0 (#298)
| Benchmark | Amazon Nova Lite | GPT-3.5-turbo |
|---|---|---|
| LMArena Expert | 1201 | 1070 |
| MMLU | 77% | 71.4% |
| GPQA Diamond | — | 28% |
| Humanity's Last Exam | 3.6% | — |
| MMLU-Pro | 60% | — |
| Vectara Hallucination Rate | 6.1% | — |
| GPQA (HELM) | 39.7% | — |
| ARC (AI2) Challenge | — | 87.4% |
| BoolQ | — | 87% |
| OpenBookQA | — | 86% |
| TriviaQA | — | 85.8% |
Multimodal Not comparable
Amazon Nova Lite: 25.0 (#118), GPT-3.5-turbo: —
| Benchmark | Amazon Nova Lite | GPT-3.5-turbo |
|---|---|---|
| LMArena Vision | 990 | — |
Multilingual Amazon Nova Lite leads
Amazon Nova Lite: 38.2 (#227), GPT-3.5-turbo: 31.6 (#253)
| Benchmark | Amazon Nova Lite | GPT-3.5-turbo |
|---|---|---|
| LMArena Non-English | 1208 | 1108 |
| LMArena Chinese | 1225 | 1075 |
| LMArena French | 1237 | 1117 |
| LMArena German | 1229 | 1090 |
| LMArena Japanese | 1153 | 1043 |
| LMArena Korean | 1153 | 1019 |
| LMArena Russian | 1216 | 1123 |
| LMArena Spanish | 1225 | 1120 |
Instruction Following Amazon Nova Lite leads
Amazon Nova Lite: 59.6 (#250), GPT-3.5-turbo: 58.1 (#257)
| Benchmark | Amazon Nova Lite | GPT-3.5-turbo |
|---|---|---|
| LMArena Instruction Following | 1205 | 1119 |
| LiveBench Instruction Following | 54.1% | — |
| IFEval | 77.6% | — |
Long Context Amazon Nova Lite leads
Amazon Nova Lite: 37.4 (#212), GPT-3.5-turbo: 34.0 (#249)
| Benchmark | Amazon Nova Lite | GPT-3.5-turbo |
|---|---|---|
| LMArena Longer Query | 1234 | 1121 |
Writing & Preference Amazon Nova Lite leads
Amazon Nova Lite: 42.3 (#232), GPT-3.5-turbo: 25.4 (#300)
| Benchmark | Amazon Nova Lite | GPT-3.5-turbo |
|---|---|---|
| LMArena Text | 1229 | 1125 |
| LMArena Creative Writing | 1197 | 1092 |
| LMArena Multi-Turn | 1199 | 1117 |
| EQ-Bench Creative Writing | — | 451 |
| WildBench | 75% | — |
| LiveBench Language | 25.9% | — |
Frequently asked questions
Is Amazon Nova Lite better than GPT-3.5-turbo?
Amazon Nova Lite is the stronger model overall, scoring 31.4 to 23.1 on the Noometry Index.
Which is cheaper, Amazon Nova Lite or GPT-3.5-turbo?
Amazon Nova Lite is cheaper. It lists at $0.06 per million input tokens and $0.24 per million output tokens; GPT-3.5-turbo lists at $0.50 and $1.50.
Is Amazon Nova Lite or GPT-3.5-turbo better for coding?
Amazon Nova Lite scores higher on coding benchmarks: 32.0 versus 23.5 in the Noometry coding category.
Which has the bigger context window?
Amazon Nova Lite does, with 300K tokens against 16K.
How many benchmarks do Amazon Nova Lite and GPT-3.5-turbo share?
18 benchmarks have published results for both models. Amazon Nova Lite has 34 scored results on Noometry and GPT-3.5-turbo has 44.