# Amazon Nova Pro vs Gemini 3.8 Flash

> Gemini 3.8 Flash is the stronger model overall, scoring 61.8 to 31.0 on the Noometry Index.

- Canonical page: https://noometry.com/compare/amazon-nova-pro-vs-gemini-3-8-flash
- Last updated: 2026-10-10
- Shared benchmarks: 20

## Summary

- They share 20 benchmarks with published results for both. Amazon Nova Pro scores higher in 0 categories and Gemini 3.8 Flash in 10 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Gemini 3.8 Flash leads 76.9 to 20.0.
- The biggest single-benchmark swing is Humanity's Last Exam: 4.4% for Amazon Nova Pro and 44.5% for Gemini 3.8 Flash.
- Amazon Nova Pro is cheaper at $0.80 / $3.20 per million input/output tokens, against $0.75 / $3.75 for Gemini 3.8 Flash.
- Gemini 3.8 Flash accepts more context: 1.05M tokens versus 300K.

## Snapshot

| | Amazon Nova Pro | Gemini 3.8 Flash |
|---|---|---|
| Provider | Amazon | Google |
| Noometry Index | 31.0 | 61.8 |
| Rank | 281 | 11 |
| Context | 300K | 1.05M |
| Input $/M | $0.80 | $0.75 |
| Output $/M | $3.20 | $3.75 |
| Weights | Proprietary | Proprietary |

## Coding

- Amazon Nova Pro: 35.1 (#229)
- Gemini 3.8 Flash: 59.2 (#15)

| Benchmark | Amazon Nova Pro | Gemini 3.8 Flash |
|---|---|---|
| LMArena Coding | 1270 | 1510 |
| DeepSWE | — | 73.8% |
| FrontierCode | — | 41.2% |
| CursorBench | — | 39.6% |
| LMArena WebDev | — | 1584 |
| FrontierSWE | — | 19.6% |
| SciCode | — | 56.6% |
| WeirdML | — | 84.8% |
| LiveBench Coding | 38.1% | — |
| ALE-Bench | — | 1,270 |

## Agentic & Tool Use

- Amazon Nova Pro: 16.7 (#147)
- Gemini 3.8 Flash: 41.8 (#21)

| Benchmark | Amazon Nova Pro | Gemini 3.8 Flash |
|---|---|---|
| APEX-Agents | — | 64.3% |
| Berkeley Function Calling Leaderboard | 25% | — |
| Remote Labor Index | — | 5.8% |
| TheAgentCompany | 1.7% | — |
| GDP.pdf | — | 23.4% |
| Vending-Bench 2 | — | 5,094 |

## Reasoning

- Amazon Nova Pro: 20.0 (#243)
- Gemini 3.8 Flash: 76.9 (#5)

| Benchmark | Amazon Nova Pro | Gemini 3.8 Flash |
|---|---|---|
| LMArena Hard Prompts | 1246 | 1508 |
| Epoch Capabilities Index | 123.8 | 156.71 |
| ARC-AGI-2 | — | 89.2% |
| NYT Connections (extended) | — | 97.4% |
| ARC-AGI-1 | — | 98.5% |
| CritPt | — | 18.3% |
| Chess Puzzles | — | 61% |
| LiveBench Reasoning | 32.6% | — |
| Mystery Game Puzzles | — | 47% |
| DTBench | — | 95.7% |
| LiveBench Data Analysis | 48.3% | — |
| LMCA | — | 52.9% |
| Surface Evolver Bench | — | 76.9% |
| LiveBench | 43.5% | — |

## Math

- Amazon Nova Pro: 28.5 (#243)
- Gemini 3.8 Flash: 65.3 (#28)

| Benchmark | Amazon Nova Pro | Gemini 3.8 Flash |
|---|---|---|
| LMArena Math | 1252 | 1528 |
| FrontierMath (Tiers 1-3) | — | 68.4% |
| FrontierMath Tier 4 | — | 22% |
| OTIS Mock AIME 2024-2025 | — | 98.9% |
| ProofBench | — | 48% |
| Omni-MATH | 24.2% | — |
| LiveBench Math | 38% | — |

## Knowledge

- Amazon Nova Pro: 27.4 (#250)
- Gemini 3.8 Flash: 74.8 (#2)

| Benchmark | Amazon Nova Pro | Gemini 3.8 Flash |
|---|---|---|
| Humanity's Last Exam | 4.4% | 44.5% |
| LMArena Expert | 1211 | 1524 |
| GPQA Diamond | — | 95.4% |
| SimpleQA Verified | — | 69.7% |
| MMLU-Pro | 67.3% | — |
| Confabulations | 30.1% | — |
| Vectara Hallucination Rate | 5.1% | — |
| GPQA (HELM) | 44.6% | — |
| MMLU | 82% | — |

## Multimodal

- Amazon Nova Pro: 25.0 (#126)
- Gemini 3.8 Flash: 40.7 (#45)

| Benchmark | Amazon Nova Pro | Gemini 3.8 Flash |
|---|---|---|
| LMArena Vision | 980 | 1314 |
| Blueprint-Bench 2 | — | 38.6% |
| Furniture Assembly | — | 31.7% |

## Multilingual

- Amazon Nova Pro: 39.7 (#223)
- Gemini 3.8 Flash: 58.0 (#5)

| Benchmark | Amazon Nova Pro | Gemini 3.8 Flash |
|---|---|---|
| LMArena Non-English | 1234 | 1491 |
| LMArena Chinese | 1244 | 1554 |
| LMArena French | 1271 | 1498 |
| LMArena German | 1243 | 1493 |
| LMArena Japanese | 1200 | 1502 |
| LMArena Korean | 1203 | 1459 |
| LMArena Russian | 1240 | 1515 |
| LMArena Spanish | 1182 | 1485 |

## Instruction Following

- Amazon Nova Pro: 64.9 (#226)
- Gemini 3.8 Flash: 78.0 (#13)

| Benchmark | Amazon Nova Pro | Gemini 3.8 Flash |
|---|---|---|
| LMArena Instruction Following | 1235 | 1490 |
| LiveBench Instruction Following | 67.1% | — |
| IFEval | 81.5% | — |

## Long Context

- Amazon Nova Pro: 38.1 (#205)
- Gemini 3.8 Flash: 46.3 (#24)

| Benchmark | Amazon Nova Pro | Gemini 3.8 Flash |
|---|---|---|
| LMArena Longer Query | 1255 | 1508 |

## Writing & Preference

- Amazon Nova Pro: 43.9 (#226)
- Gemini 3.8 Flash: 72.2 (#15)

| Benchmark | Amazon Nova Pro | Gemini 3.8 Flash |
|---|---|---|
| LMArena Text | 1259 | 1499 |
| LMArena Creative Writing | 1212 | 1492 |
| LMArena Multi-Turn | 1246 | 1501 |
| Short-Story Creative Writing | 60.5% | — |
| EQ-Bench Creative Writing | — | 1748 |
| WildBench | 77.7% | — |
| LiveBench Language | 37% | — |

## FAQ

### Is Amazon Nova Pro better than Gemini 3.8 Flash?

Gemini 3.8 Flash is the stronger model overall, scoring 61.8 to 31.0 on the Noometry Index.

### Which is cheaper, Amazon Nova Pro or Gemini 3.8 Flash?

Amazon Nova Pro is cheaper. It lists at $0.80 per million input tokens and $3.20 per million output tokens; Gemini 3.8 Flash lists at $0.75 and $3.75.

### Is Amazon Nova Pro or Gemini 3.8 Flash better for coding?

Gemini 3.8 Flash scores higher on coding benchmarks: 59.2 versus 35.1 in the Noometry coding category.

### Which has the bigger context window?

Gemini 3.8 Flash does, with 1.05M tokens against 300K.

### How many benchmarks do Amazon Nova Pro and Gemini 3.8 Flash share?

20 benchmarks have published results for both models. Amazon Nova Pro has 38 scored results on Noometry and Gemini 3.8 Flash has 50.