# Amazon Nova Lite vs DeepSeek V4 Pro

> DeepSeek V4 Pro is the stronger model overall, scoring 54.3 to 31.9 on the Noometry Index. Amazon Nova Lite costs 9.4× less per token, which makes it the better buy when DeepSeek V4 Pro's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/amazon-nova-lite-vs-deepseek-v4-pro
- Last updated: 2026-10-10
- Shared benchmarks: 19

## Summary

- They share 19 benchmarks with published results for both. Amazon Nova Lite scores higher in 0 categories and DeepSeek V4 Pro in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where DeepSeek V4 Pro leads 56.5 to 19.3.
- Amazon Nova Lite is cheaper at $0.06 / $0.24 per million input/output tokens, against $0.66 / $1.98 for DeepSeek V4 Pro.
- DeepSeek V4 Pro accepts more context: 1M tokens versus 300K.
- DeepSeek V4 Pro has downloadable open weights; the other is API-only.

## Snapshot

| | Amazon Nova Lite | DeepSeek V4 Pro |
|---|---|---|
| Provider | Amazon | DeepSeek |
| Noometry Index | 31.9 | 54.3 |
| Rank | 265 | 31 |
| Context | 300K | 1M |
| Input $/M | $0.06 | $0.66 |
| Output $/M | $0.24 | $1.98 |
| Weights | Proprietary | Open |

## Coding

- Amazon Nova Lite: 32.5 (#270)
- DeepSeek V4 Pro: 52.4 (#34)

| Benchmark | Amazon Nova Lite | DeepSeek V4 Pro |
|---|---|---|
| LMArena Coding | 1239 | 1470 |
| ALE-Bench | 236.25 | 1,403 |
| SWE-bench Verified | — | 77.6% |
| FrontierCode | — | 28.6% |
| LMArena WebDev | — | 1582 |
| SciCode | — | 51% |
| WeirdML | — | 66.2% |
| LiveBench Coding | 27.5% | — |

## Agentic & Tool Use

- Amazon Nova Lite: —
- DeepSeek V4 Pro: 32.8 (#58)

| Benchmark | Amazon Nova Lite | DeepSeek V4 Pro |
|---|---|---|
| APEX-Agents | — | 47.3% |
| Vending-Bench 2 | — | 3,285 |

## Reasoning

- Amazon Nova Lite: 19.3 (#260)
- DeepSeek V4 Pro: 56.5 (#24)

| Benchmark | Amazon Nova Lite | DeepSeek V4 Pro |
|---|---|---|
| LMArena Hard Prompts | 1220 | 1461 |
| ARC-AGI-2 | — | 61.3% |
| Kagi LLM Benchmark | — | 53.5% |
| NYT Connections (extended) | — | 91.3% |
| ARC-AGI-1 | — | 90.5% |
| CritPt | — | 18% |
| Chess Puzzles | — | 47% |
| LiveBench Reasoning | 36.7% | — |
| Mystery Game Puzzles | — | 43% |
| DTBench | — | 93.9% |
| LiveBench Data Analysis | 37.2% | — |
| LMCA | — | 45.5% |
| Surface Evolver Bench | — | 40% |
| Epoch Capabilities Index | — | 155.31 |
| ForecastBench | — | 56.1 |
| LiveBench | 36.4% | — |

## Math

- Amazon Nova Lite: 27.8 (#247)
- DeepSeek V4 Pro: 64.8 (#30)

| Benchmark | Amazon Nova Lite | DeepSeek V4 Pro |
|---|---|---|
| LMArena Math | 1227 | 1455 |
| FrontierMath (Tiers 1-3) | — | 64.6% |
| FrontierMath Tier 4 | — | 26.8% |
| MathArena Final-Answer Competitions | — | 76.6% |
| OTIS Mock AIME 2024-2025 | — | 98.6% |
| ProofBench | — | 50% |
| Omni-MATH | 23.3% | — |
| LiveBench Math | 36.7% | — |

## Knowledge

- Amazon Nova Lite: 26.3 (#259)
- DeepSeek V4 Pro: 59.5 (#31)

| Benchmark | Amazon Nova Lite | DeepSeek V4 Pro |
|---|---|---|
| Vectara Hallucination Rate | 6.1% | 8.6% |
| LMArena Expert | 1201 | 1464 |
| GPQA Diamond | — | 91.7% |
| Humanity's Last Exam | 3.6% | — |
| SimpleQA Verified | — | 52.9% |
| MMLU-Pro | 60% | — |
| GPQA (HELM) | 39.7% | — |
| MMLU | 77% | — |

## Multimodal

- Amazon Nova Lite: 25.5 (#123)
- DeepSeek V4 Pro: —

| Benchmark | Amazon Nova Lite | DeepSeek V4 Pro |
|---|---|---|
| LMArena Vision | 990 | — |

## Multilingual

- Amazon Nova Lite: 38.0 (#232)
- DeepSeek V4 Pro: 54.4 (#45)

| Benchmark | Amazon Nova Lite | DeepSeek V4 Pro |
|---|---|---|
| LMArena Non-English | 1208 | 1439 |
| LMArena Chinese | 1225 | 1486 |
| LMArena French | 1237 | 1472 |
| LMArena German | 1229 | 1458 |
| LMArena Japanese | 1153 | 1445 |
| LMArena Korean | 1153 | 1447 |
| LMArena Russian | 1216 | 1453 |
| LMArena Spanish | 1226 | 1458 |

## Instruction Following

- Amazon Nova Lite: 59.3 (#255)
- DeepSeek V4 Pro: 76.1 (#47)

| Benchmark | Amazon Nova Lite | DeepSeek V4 Pro |
|---|---|---|
| LMArena Instruction Following | 1205 | 1448 |
| LiveBench Instruction Following | 54.1% | — |
| IFEval | 77.6% | — |

## Long Context

- Amazon Nova Lite: 37.4 (#217)
- DeepSeek V4 Pro: 45.0 (#51)

| Benchmark | Amazon Nova Lite | DeepSeek V4 Pro |
|---|---|---|
| LMArena Longer Query | 1234 | 1458 |
| CL-bench Life | — | 13.5% |

## Writing & Preference

- Amazon Nova Lite: 42.1 (#237)
- DeepSeek V4 Pro: 65.5 (#46)

| Benchmark | Amazon Nova Lite | DeepSeek V4 Pro |
|---|---|---|
| LMArena Text | 1229 | 1451 |
| LMArena Creative Writing | 1197 | 1446 |
| LMArena Multi-Turn | 1199 | 1467 |
| EQ-Bench Creative Writing | — | 1553 |
| WildBench | 75% | — |
| EQ-Bench 4 | — | 1166 |
| LiveBench Language | 25.9% | — |

## FAQ

### Is Amazon Nova Lite better than DeepSeek V4 Pro?

DeepSeek V4 Pro is the stronger model overall, scoring 54.3 to 31.9 on the Noometry Index. Amazon Nova Lite costs 9.4× less per token, which makes it the better buy when DeepSeek V4 Pro's lead doesn't matter for your workload.

### Which is cheaper, Amazon Nova Lite or DeepSeek V4 Pro?

Amazon Nova Lite is cheaper. It lists at $0.06 per million input tokens and $0.24 per million output tokens; DeepSeek V4 Pro lists at $0.66 and $1.98.

### Is Amazon Nova Lite or DeepSeek V4 Pro better for coding?

DeepSeek V4 Pro scores higher on coding benchmarks: 52.4 versus 32.5 in the Noometry coding category.

### Which has the bigger context window?

DeepSeek V4 Pro does, with 1M tokens against 300K.

### How many benchmarks do Amazon Nova Lite and DeepSeek V4 Pro share?

19 benchmarks have published results for both models. Amazon Nova Lite has 34 scored results on Noometry and DeepSeek V4 Pro has 48.