Model comparison
Amazon Nova Micro vs Gemma 3 12B
Gemma 3 12B is the stronger model overall, scoring 31.8 to 29.9 on the Noometry Index.
Last verified . 14 shared benchmarks.
GoogleSummary
- They share 14 benchmarks with published results for both. Amazon Nova Micro scores higher in 3 categories and Gemma 3 12B in 6 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in instruction following, where Gemma 3 12B leads 68.8 to 56.6.
- The biggest single-benchmark swing is Berkeley Function Calling Leaderboard: 22.3% for Amazon Nova Micro and 30.4% for Gemma 3 12B.
- Amazon Nova Micro is cheaper at $0.035 / $0.14 per million input/output tokens, against $0.05 / $0.15 for Gemma 3 12B.
- Gemma 3 12B accepts more context: 131K tokens versus 128K.
- Gemma 3 12B has downloadable open weights; the other is API-only.
Side by side
| Amazon Nova Micro | Gemma 3 12B | |
|---|---|---|
| Provider | Amazon | |
| Noometry Index | 29.9 | 31.8 |
| Released | 2024-12-03 | 2025-03-12 |
| Weights | Proprietary | Open |
| Context window | 128K | 131K |
| Max output | 10K | 8K |
| Input $ / M tokens | $0.035 | $0.05 |
| Output $ / M tokens | $0.14 | $0.15 |
| Results tracked | 32 | 24 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemma 3 12B leads
Amazon Nova Micro: 30.0 (#289), Gemma 3 12B: 31.2 (#276)
| Benchmark | Amazon Nova Micro | Gemma 3 12B |
|---|---|---|
| LMArena Coding | 1218 | 1281 |
| SciCode | — | 17.4% |
| LiveBench Coding | 20.2% | — |
Agentic & Tool Use Gemma 3 12B leads
Amazon Nova Micro: 21.9 (#125), Gemma 3 12B: 25.3 (#101)
| Benchmark | Amazon Nova Micro | Gemma 3 12B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 22.3% | 30.4% |
Reasoning Amazon Nova Micro leads
Amazon Nova Micro: 16.7 (#289), Gemma 3 12B: 15.2 (#307)
| Benchmark | Amazon Nova Micro | Gemma 3 12B |
|---|---|---|
| LMArena Hard Prompts | 1191 | 1309 |
| CritPt | — | 0% |
| Chess Puzzles | — | 0% |
| LiveBench Reasoning | 25.1% | — |
| DTBench | — | 48.8% |
| LiveBench Data Analysis | 34% | — |
| LMCA | — | 4.5% |
| Epoch Capabilities Index | — | 123.49 |
| LiveBench | 29.6% | — |
Math Amazon Nova Micro leads
Amazon Nova Micro: 24.9 (#252), Gemma 3 12B: 21.0 (#273)
| Benchmark | Amazon Nova Micro | Gemma 3 12B |
|---|---|---|
| LMArena Math | 1206 | 1307 |
| OTIS Mock AIME 2024-2025 | — | 16.7% |
| Omni-MATH | 21.4% | — |
| LiveBench Math | 34.5% | — |
Knowledge Amazon Nova Micro leads
Amazon Nova Micro: 29.2 (#231), Gemma 3 12B: 26.5 (#250)
| Benchmark | Amazon Nova Micro | Gemma 3 12B |
|---|---|---|
| Vectara Hallucination Rate | 5.5% | 4.4% |
| LMArena Expert | 1184 | 1248 |
| GPQA Diamond | — | 39.5% |
| MMLU-Pro | 51.1% | — |
| GPQA (HELM) | 38.3% | — |
| MMLU | 70.8% | — |
Multimodal Not comparable
Amazon Nova Micro: —, Gemma 3 12B: —
| Benchmark | Amazon Nova Micro | Gemma 3 12B |
|---|---|---|
| MindCube | — | 46.7% |
Multilingual Gemma 3 12B leads
Amazon Nova Micro: 36.7 (#234), Gemma 3 12B: 45.9 (#160)
| Benchmark | Amazon Nova Micro | Gemma 3 12B |
|---|---|---|
| LMArena Non-English | 1186 | 1318 |
| LMArena German | 1192 | 1370 |
| LMArena Russian | 1185 | 1335 |
| LMArena Chinese | 1209 | — |
| LMArena French | 1237 | — |
| LMArena Japanese | 1154 | — |
| LMArena Korean | 1150 | — |
| LMArena Spanish | 1225 | — |
Instruction Following Gemma 3 12B leads
Amazon Nova Micro: 56.6 (#267), Gemma 3 12B: 68.8 (#179)
| Benchmark | Amazon Nova Micro | Gemma 3 12B |
|---|---|---|
| LMArena Instruction Following | 1174 | 1299 |
| LiveBench Instruction Following | 48% | — |
| IFEval | 76% | — |
Long Context Gemma 3 12B leads
Amazon Nova Micro: 36.5 (#224), Gemma 3 12B: 40.0 (#157)
| Benchmark | Amazon Nova Micro | Gemma 3 12B |
|---|---|---|
| LMArena Longer Query | 1206 | 1317 |
Writing & Preference Gemma 3 12B leads
Amazon Nova Micro: 39.7 (#242), Gemma 3 12B: 47.7 (#205)
| Benchmark | Amazon Nova Micro | Gemma 3 12B |
|---|---|---|
| LMArena Text | 1208 | 1334 |
| LMArena Creative Writing | 1172 | 1331 |
| LMArena Multi-Turn | 1178 | 1334 |
| EQ-Bench Creative Writing | — | 1126 |
| WildBench | 74.3% | — |
| LiveBench Language | 15.8% | — |
Frequently asked questions
Is Amazon Nova Micro better than Gemma 3 12B?
Gemma 3 12B is the stronger model overall, scoring 31.8 to 29.9 on the Noometry Index.
Which is cheaper, Amazon Nova Micro or Gemma 3 12B?
Amazon Nova Micro is cheaper. It lists at $0.035 per million input tokens and $0.14 per million output tokens; Gemma 3 12B lists at $0.05 and $0.15.
Is Amazon Nova Micro or Gemma 3 12B better for coding?
Gemma 3 12B scores higher on coding benchmarks: 31.2 versus 30.0 in the Noometry coding category.
Which has the bigger context window?
Gemma 3 12B does, with 131K tokens against 128K.
How many benchmarks do Amazon Nova Micro and Gemma 3 12B share?
14 benchmarks have published results for both models. Amazon Nova Micro has 32 scored results on Noometry and Gemma 3 12B has 24.