Model comparison
Amazon Nova Pro vs Gemma 4 31B IT
Gemma 4 31B IT is the stronger model overall, scoring 43.5 to 31.0 on the Noometry Index.
Last verified . 17 shared benchmarks.
GoogleSummary
- They share 17 benchmarks with published results for both. Amazon Nova Pro scores higher in 0 categories and Gemma 4 31B IT in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in multimodal, where Gemma 4 31B IT leads 41.6 to 25.0.
- Gemma 4 31B IT is cheaper at $0.09 / $0.34 per million input/output tokens, against $0.80 / $3.20 for Amazon Nova Pro.
- Amazon Nova Pro accepts more context: 300K tokens versus 262K.
- Gemma 4 31B IT has downloadable open weights; the other is API-only.
Side by side
| Amazon Nova Pro | Gemma 4 31B IT | |
|---|---|---|
| Provider | Amazon | |
| Noometry Index | 31.0 | 43.5 |
| Released | 2024-12-03 | 2026-04-02 |
| Weights | Proprietary | Open |
| Context window | 300K | 262K |
| Max output | 10K | 33K |
| Input $ / M tokens | $0.80 | $0.09 |
| Output $ / M tokens | $3.20 | $0.34 |
| Results tracked | 38 | 35 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemma 4 31B IT leads
Amazon Nova Pro: 35.1 (#229), Gemma 4 31B IT: 42.3 (#108)
| Benchmark | Amazon Nova Pro | Gemma 4 31B IT |
|---|---|---|
| LMArena Coding | 1270 | 1459 |
| LMArena WebDev | — | 1366 |
| SciCode | — | 43.4% |
| WeirdML | — | 52.3% |
| LiveBench Coding | 38.1% | — |
| ALE-Bench | — | 925.5 |
Agentic & Tool Use Not comparable
Amazon Nova Pro: 16.7 (#147), Gemma 4 31B IT: —
| Benchmark | Amazon Nova Pro | Gemma 4 31B IT |
|---|---|---|
| Berkeley Function Calling Leaderboard | 25% | — |
| TheAgentCompany | 1.7% | — |
Reasoning Gemma 4 31B IT leads
Amazon Nova Pro: 20.0 (#243), Gemma 4 31B IT: 27.2 (#122)
| Benchmark | Amazon Nova Pro | Gemma 4 31B IT |
|---|---|---|
| LMArena Hard Prompts | 1246 | 1448 |
| Epoch Capabilities Index | 123.8 | 142.74 |
| Kagi LLM Benchmark | — | 63.5% |
| NYT Connections (extended) | — | 70.6% |
| CritPt | — | 1.4% |
| Chess Puzzles | — | 5% |
| Thematic Generalization | — | 53% |
| LiveBench Reasoning | 32.6% | — |
| DTBench | — | 82.7% |
| LiveBench Data Analysis | 48.3% | — |
| LMCA | — | 39.3% |
| Surface Evolver Bench | — | 30.6% |
| LiveBench | 43.5% | — |
Math Gemma 4 31B IT leads
Amazon Nova Pro: 28.5 (#243), Gemma 4 31B IT: 43.2 (#81)
| Benchmark | Amazon Nova Pro | Gemma 4 31B IT |
|---|---|---|
| LMArena Math | 1252 | 1465 |
| OTIS Mock AIME 2024-2025 | — | 73.3% |
| Omni-MATH | 24.2% | — |
| LiveBench Math | 38% | — |
Knowledge Gemma 4 31B IT leads
Amazon Nova Pro: 27.4 (#250), Gemma 4 31B IT: 37.9 (#151)
| Benchmark | Amazon Nova Pro | Gemma 4 31B IT |
|---|---|---|
| Vectara Hallucination Rate | 5.1% | 7.4% |
| LMArena Expert | 1211 | 1465 |
| GPQA Diamond | — | 75.8% |
| Humanity's Last Exam | 4.4% | — |
| SimpleQA Verified | — | 10.4% |
| MMLU-Pro | 67.3% | — |
| Confabulations | 30.1% | — |
| GPQA (HELM) | 44.6% | — |
| MMLU | 82% | — |
Multimodal Gemma 4 31B IT leads
Amazon Nova Pro: 25.0 (#126), Gemma 4 31B IT: 41.6 (#34)
| Benchmark | Amazon Nova Pro | Gemma 4 31B IT |
|---|---|---|
| LMArena Vision | 980 | 1277 |
| LMArena Document | — | 1425 |
Multilingual Gemma 4 31B IT leads
Amazon Nova Pro: 39.7 (#223), Gemma 4 31B IT: 53.8 (#57)
| Benchmark | Amazon Nova Pro | Gemma 4 31B IT |
|---|---|---|
| LMArena Non-English | 1234 | 1431 |
| LMArena Chinese | 1244 | 1476 |
| LMArena French | 1271 | 1435 |
| LMArena Russian | 1240 | 1460 |
| LMArena Spanish | 1182 | 1444 |
| LMArena German | 1243 | — |
| LMArena Japanese | 1200 | — |
| LMArena Korean | 1203 | — |
Instruction Following Gemma 4 31B IT leads
Amazon Nova Pro: 64.9 (#226), Gemma 4 31B IT: 75.5 (#61)
| Benchmark | Amazon Nova Pro | Gemma 4 31B IT |
|---|---|---|
| LMArena Instruction Following | 1235 | 1433 |
| LiveBench Instruction Following | 67.1% | — |
| IFEval | 81.5% | — |
Long Context Gemma 4 31B IT leads
Amazon Nova Pro: 38.1 (#205), Gemma 4 31B IT: 44.2 (#71)
| Benchmark | Amazon Nova Pro | Gemma 4 31B IT |
|---|---|---|
| LMArena Longer Query | 1255 | 1446 |
Writing & Preference Gemma 4 31B IT leads
Amazon Nova Pro: 43.9 (#226), Gemma 4 31B IT: 60.5 (#96)
| Benchmark | Amazon Nova Pro | Gemma 4 31B IT |
|---|---|---|
| LMArena Text | 1259 | 1443 |
| LMArena Creative Writing | 1212 | 1415 |
| LMArena Multi-Turn | 1246 | 1452 |
| Short-Story Creative Writing | 60.5% | — |
| EQ-Bench Creative Writing | — | 1368 |
| WildBench | 77.7% | — |
| EQ-Bench 4 | — | 1120 |
| LiveBench Language | 37% | — |
Frequently asked questions
Is Amazon Nova Pro better than Gemma 4 31B IT?
Gemma 4 31B IT is the stronger model overall, scoring 43.5 to 31.0 on the Noometry Index.
Which is cheaper, Amazon Nova Pro or Gemma 4 31B IT?
Gemma 4 31B IT is cheaper. It lists at $0.09 per million input tokens and $0.34 per million output tokens; Amazon Nova Pro lists at $0.80 and $3.20.
Is Amazon Nova Pro or Gemma 4 31B IT better for coding?
Gemma 4 31B IT scores higher on coding benchmarks: 42.3 versus 35.1 in the Noometry coding category.
Which has the bigger context window?
Amazon Nova Pro does, with 300K tokens against 262K.
How many benchmarks do Amazon Nova Pro and Gemma 4 31B IT share?
17 benchmarks have published results for both models. Amazon Nova Pro has 38 scored results on Noometry and Gemma 4 31B IT has 35.