Model comparison
Amazon Nova Lite vs o3-mini
o3-mini is the stronger model overall, scoring 36.7 to 31.9 on the Noometry Index. Amazon Nova Lite costs 18× less per token, which makes it the better buy when o3-mini's lead doesn't matter for your workload.
Last verified . 24 shared benchmarks.
Summary
- They share 24 benchmarks with published results for both. Amazon Nova Lite scores higher in 2 categories and o3-mini in 6 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in instruction following, where o3-mini leads 75.1 to 59.3.
- The biggest single-benchmark swing is LiveBench Coding: 27.5% for Amazon Nova Lite and 82.7% for o3-mini.
- Amazon Nova Lite is cheaper at $0.06 / $0.24 per million input/output tokens, against $1.10 / $4.40 for o3-mini.
- Amazon Nova Lite accepts more context: 300K tokens versus 200K.
Side by side
| Amazon Nova Lite | o3-mini | |
|---|---|---|
| Provider | Amazon | OpenAI |
| Noometry Index | 31.9 | 36.7 |
| Released | 2024-12-03 | 2024-12-20 |
| Weights | Proprietary | Proprietary |
| Context window | 300K | 200K |
| Max output | 10K | 100K |
| Input $ / M tokens | $0.06 | $1.10 |
| Output $ / M tokens | $0.24 | $4.40 |
| Results tracked | 34 | 51 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o3-mini leads
Amazon Nova Lite: 32.5 (#270), o3-mini: 40.8 (#132)
| Benchmark | Amazon Nova Lite | o3-mini |
|---|---|---|
| LiveBench Coding | 27.5% | 82.7% |
| LMArena Coding | 1239 | 1378 |
| Aider Polyglot | — | 60.4% |
| SciCode | — | 39.8% |
| GSO | — | 1.3% |
| WeirdML | — | 43.7% |
| CadEval | — | 54% |
| ALE-Bench | 236.25 | — |
Agentic & Tool Use Not comparable
Amazon Nova Lite: —, o3-mini: 29.6 (#84)
| Benchmark | Amazon Nova Lite | o3-mini |
|---|---|---|
| Cybench | — | 22.5% |
Reasoning Amazon Nova Lite leads
Amazon Nova Lite: 19.3 (#260), o3-mini: 16.3 (#305)
| Benchmark | Amazon Nova Lite | o3-mini |
|---|---|---|
| LiveBench Reasoning | 36.7% | 89.6% |
| LMArena Hard Prompts | 1220 | 1366 |
| LiveBench Data Analysis | 37.2% | 70.6% |
| LiveBench | 36.4% | 75.9% |
| ARC-AGI-2 | — | 3% |
| SimpleBench | — | 22.8% |
| ARC-AGI-1 | — | 34.5% |
| CritPt | — | 0.3% |
| Chess Puzzles | — | 17% |
| Mystery Game Puzzles | — | 7% |
| DTBench | — | 68.8% |
| LMCA | — | 19% |
| Epoch Capabilities Index | — | 140.34 |
| ForecastBench | — | 59.6 |
Math Too close to call
Amazon Nova Lite: 27.8 (#247), o3-mini: 28.1 (#244)
| Benchmark | Amazon Nova Lite | o3-mini |
|---|---|---|
| LiveBench Math | 36.7% | 77.3% |
| LMArena Math | 1227 | 1396 |
| FrontierMath (Tiers 1-3) | — | 18.6% |
| FrontierMath Tier 4 | — | 0% |
| OTIS Mock AIME 2024-2025 | — | 76.9% |
| Omni-MATH | 23.3% | — |
| MATH Level 5 | — | 96.5% |
| FrontierMath (Feb 2025 set) | — | 12.4% |
| FrontierMath Tier 4 (v1) | — | 4.2% |
Knowledge o3-mini leads
Amazon Nova Lite: 26.3 (#259), o3-mini: 38.3 (#146)
| Benchmark | Amazon Nova Lite | o3-mini |
|---|---|---|
| LMArena Expert | 1201 | 1364 |
| GPQA Diamond | — | 77% |
| Humanity's Last Exam | 3.6% | — |
| SimpleQA Verified | — | 15.3% |
| MMLU-Pro | 60% | — |
| Confabulations | — | 17.9% |
| Vectara Hallucination Rate | 6.1% | — |
| GPQA (HELM) | 39.7% | — |
| MMLU | 77% | — |
Multimodal Not comparable
Amazon Nova Lite: 25.5 (#123), o3-mini: —
| Benchmark | Amazon Nova Lite | o3-mini |
|---|---|---|
| LMArena Vision | 990 | — |
Multilingual o3-mini leads
Amazon Nova Lite: 38.0 (#232), o3-mini: 45.7 (#164)
| Benchmark | Amazon Nova Lite | o3-mini |
|---|---|---|
| LMArena Non-English | 1208 | 1319 |
| LMArena Chinese | 1225 | 1379 |
| LMArena French | 1237 | 1334 |
| LMArena German | 1229 | 1303 |
| LMArena Japanese | 1153 | 1286 |
| LMArena Korean | 1153 | 1314 |
| LMArena Russian | 1216 | 1304 |
| LMArena Spanish | 1226 | 1321 |
Instruction Following o3-mini leads
Amazon Nova Lite: 59.3 (#255), o3-mini: 75.1 (#72)
| Benchmark | Amazon Nova Lite | o3-mini |
|---|---|---|
| LiveBench Instruction Following | 54.1% | 84.4% |
| LMArena Instruction Following | 1205 | 1337 |
| IFEval | 77.6% | — |
Long Context Amazon Nova Lite leads
Amazon Nova Lite: 37.4 (#217), o3-mini: 33.8 (#256)
| Benchmark | Amazon Nova Lite | o3-mini |
|---|---|---|
| LMArena Longer Query | 1234 | 1343 |
| Fiction.LiveBench | — | 50% |
Writing & Preference o3-mini leads
Amazon Nova Lite: 42.1 (#237), o3-mini: 50.3 (#182)
| Benchmark | Amazon Nova Lite | o3-mini |
|---|---|---|
| LMArena Text | 1229 | 1337 |
| LMArena Creative Writing | 1197 | 1286 |
| LMArena Multi-Turn | 1199 | 1320 |
| LiveBench Language | 25.9% | 50.7% |
| Short-Story Creative Writing | — | 61.7% |
| WildBench | 75% | — |
Frequently asked questions
Is Amazon Nova Lite better than o3-mini?
o3-mini is the stronger model overall, scoring 36.7 to 31.9 on the Noometry Index. Amazon Nova Lite costs 18× less per token, which makes it the better buy when o3-mini's lead doesn't matter for your workload.
Which is cheaper, Amazon Nova Lite or o3-mini?
Amazon Nova Lite is cheaper. It lists at $0.06 per million input tokens and $0.24 per million output tokens; o3-mini lists at $1.10 and $4.40.
Is Amazon Nova Lite or o3-mini better for coding?
o3-mini scores higher on coding benchmarks: 40.8 versus 32.5 in the Noometry coding category.
Which has the bigger context window?
Amazon Nova Lite does, with 300K tokens against 200K.
How many benchmarks do Amazon Nova Lite and o3-mini share?
24 benchmarks have published results for both models. Amazon Nova Lite has 34 scored results on Noometry and o3-mini has 51.