Model comparison
Amazon Nova Lite vs Gemma 2 27B
Amazon Nova Lite is the stronger model overall, scoring 31.4 to 29.2 on the Noometry Index.
Last verified . 25 shared benchmarks.
GoogleSummary
- They share 25 benchmarks with published results for both. Amazon Nova Lite scores higher in 4 categories and Gemma 2 27B in 4 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in math, where Amazon Nova Lite leads 25.8 to 10.0.
- The biggest single-benchmark swing is LiveBench Data Analysis: 37.2% for Amazon Nova Lite and 47.9% for Gemma 2 27B.
- Amazon Nova Lite is cheaper at $0.06 / $0.24 per million input/output tokens, against $0.65 / $0.65 for Gemma 2 27B.
- Amazon Nova Lite accepts more context: 300K tokens versus 8K.
- Gemma 2 27B has downloadable open weights; the other is API-only.
Side by side
| Amazon Nova Lite | Gemma 2 27B | |
|---|---|---|
| Provider | Amazon | |
| Noometry Index | 31.4 | 29.2 |
| Released | 2024-12-03 | 2024-06-24 |
| Weights | Proprietary | Open |
| Context window | 300K | 8K |
| Max output | 10K | 2K |
| Input $ / M tokens | $0.06 | $0.65 |
| Output $ / M tokens | $0.24 | $0.65 |
| Results tracked | 34 | 34 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemma 2 27B leads
Amazon Nova Lite: 32.0 (#266), Gemma 2 27B: 33.5 (#243)
| Benchmark | Amazon Nova Lite | Gemma 2 27B |
|---|---|---|
| LiveBench Coding | 27.5% | 36% |
| LMArena Coding | 1239 | 1211 |
| BigCodeBench Instruct | — | 42.8% |
| BigCodeBench Complete | — | 52.5% |
| ALE-Bench | 236.25 | — |
Reasoning Amazon Nova Lite leads
Amazon Nova Lite: 18.5 (#257), Gemma 2 27B: 14.8 (#313)
| Benchmark | Amazon Nova Lite | Gemma 2 27B |
|---|---|---|
| LiveBench Reasoning | 36.7% | 28.1% |
| LMArena Hard Prompts | 1220 | 1199 |
| LiveBench Data Analysis | 37.2% | 47.9% |
| LiveBench | 36.4% | 38.2% |
| DTBench | — | 48% |
| LMCA | — | 7.1% |
| Epoch Capabilities Index | — | 122.06 |
Math Amazon Nova Lite leads
Amazon Nova Lite: 25.8 (#243), Gemma 2 27B: 10.0 (#306)
| Benchmark | Amazon Nova Lite | Gemma 2 27B |
|---|---|---|
| LiveBench Math | 36.7% | 26.5% |
| LMArena Math | 1227 | 1212 |
| OTIS Mock AIME 2024-2025 | — | 1.4% |
| Omni-MATH | 23.3% | — |
| MATH Level 5 | — | 27.9% |
Knowledge Amazon Nova Lite leads
Amazon Nova Lite: 26.0 (#254), Gemma 2 27B: 18.9 (#275)
| Benchmark | Amazon Nova Lite | Gemma 2 27B |
|---|---|---|
| LMArena Expert | 1201 | 1172 |
| MMLU | 77% | 75.7% |
| GPQA Diamond | — | 36.5% |
| Humanity's Last Exam | 3.6% | — |
| MMLU-Pro | 60% | — |
| Confabulations | — | 27.1% |
| Vectara Hallucination Rate | 6.1% | — |
| GPQA (HELM) | 39.7% | — |
Multimodal Not comparable
Amazon Nova Lite: 25.0 (#118), Gemma 2 27B: —
| Benchmark | Amazon Nova Lite | Gemma 2 27B |
|---|---|---|
| LMArena Vision | 990 | — |
Multilingual Too close to call
Amazon Nova Lite: 38.2 (#227), Gemma 2 27B: 38.8 (#221)
| Benchmark | Amazon Nova Lite | Gemma 2 27B |
|---|---|---|
| LMArena Non-English | 1208 | 1217 |
| LMArena Chinese | 1225 | 1221 |
| LMArena French | 1237 | 1246 |
| LMArena German | 1229 | 1209 |
| LMArena Japanese | 1153 | 1175 |
| LMArena Korean | 1153 | 1174 |
| LMArena Russian | 1216 | 1234 |
| LMArena Spanish | 1225 | 1228 |
Instruction Following Gemma 2 27B leads
Amazon Nova Lite: 59.6 (#250), Gemma 2 27B: 60.7 (#244)
| Benchmark | Amazon Nova Lite | Gemma 2 27B |
|---|---|---|
| LiveBench Instruction Following | 54.1% | 58.1% |
| LMArena Instruction Following | 1205 | 1206 |
| IFEval | 77.6% | — |
Long Context Too close to call
Amazon Nova Lite: 37.4 (#212), Gemma 2 27B: 37.3 (#213)
| Benchmark | Amazon Nova Lite | Gemma 2 27B |
|---|---|---|
| LMArena Longer Query | 1234 | 1231 |
Writing & Preference Gemma 2 27B leads
Amazon Nova Lite: 42.3 (#232), Gemma 2 27B: 44.4 (#220)
| Benchmark | Amazon Nova Lite | Gemma 2 27B |
|---|---|---|
| LMArena Text | 1229 | 1231 |
| LMArena Creative Writing | 1197 | 1241 |
| LMArena Multi-Turn | 1199 | 1224 |
| LiveBench Language | 25.9% | 32.6% |
| WildBench | 75% | — |
Frequently asked questions
Is Amazon Nova Lite better than Gemma 2 27B?
Amazon Nova Lite is the stronger model overall, scoring 31.4 to 29.2 on the Noometry Index.
Which is cheaper, Amazon Nova Lite or Gemma 2 27B?
Amazon Nova Lite is cheaper. It lists at $0.06 per million input tokens and $0.24 per million output tokens; Gemma 2 27B lists at $0.65 and $0.65.
Is Amazon Nova Lite or Gemma 2 27B better for coding?
Gemma 2 27B scores higher on coding benchmarks: 33.5 versus 32.0 in the Noometry coding category.
Which has the bigger context window?
Amazon Nova Lite does, with 300K tokens against 8K.
How many benchmarks do Amazon Nova Lite and Gemma 2 27B share?
25 benchmarks have published results for both models. Amazon Nova Lite has 34 scored results on Noometry and Gemma 2 27B has 34.