Noometry

Model comparison

Amazon Nova Pro vs DeepSeek-V3.1

DeepSeek-V3.1 is the stronger model overall, scoring 42.8 to 31.0 on the Noometry Index.

Last verified . 19 shared benchmarks.

Amazon Nova Pro Amazon

31.0

Rank #281 Confirmed

DeepSeek-V3.1 DeepSeek

42.8

Rank #108 Confirmed

Summary

  • They share 19 benchmarks with published results for both. Amazon Nova Pro scores higher in 1 category and DeepSeek-V3.1 in 7 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in writing & preference, where DeepSeek-V3.1 leads 60.3 to 43.9.
  • DeepSeek-V3.1 is cheaper at $0.25 / $0.95 per million input/output tokens, against $0.80 / $3.20 for Amazon Nova Pro.
  • Amazon Nova Pro accepts more context: 300K tokens versus 164K.
  • DeepSeek-V3.1 has downloadable open weights; the other is API-only.

Side by side

Amazon Nova ProDeepSeek-V3.1
ProviderAmazonDeepSeek
Noometry Index31.042.8
Released2024-12-032025-08-21
WeightsProprietaryOpen
Context window300K164K
Max output10K8K
Input $ / M tokens$0.80$0.25
Output $ / M tokens$3.20$0.95
Results tracked3827

Sponsored placements are available on pages like this one. Advertise on Noometry

Category by category

Coding DeepSeek-V3.1 leads

Amazon Nova Pro: 35.1 (#229), DeepSeek-V3.1: 40.3 (#144)

BenchmarkAmazon Nova ProDeepSeek-V3.1
LMArena Coding12701417
WeirdML—38.4%
LiveBench Coding38.1%—

Agentic & Tool Use Not comparable

Amazon Nova Pro: 16.7 (#147), DeepSeek-V3.1: —

BenchmarkAmazon Nova ProDeepSeek-V3.1
Berkeley Function Calling Leaderboard25%—
TheAgentCompany1.7%—

Reasoning DeepSeek-V3.1 leads

Amazon Nova Pro: 20.0 (#243), DeepSeek-V3.1: 27.9 (#110)

BenchmarkAmazon Nova ProDeepSeek-V3.1
LMArena Hard Prompts12461417
Epoch Capabilities Index123.8139.92
SimpleBench—40%
Kagi LLM Benchmark—53.2%
LiveBench Reasoning32.6%—
DTBench—82.7%
LiveBench Data Analysis48.3%—
LMCA—24.3%
ForecastBench—58
LiveBench43.5%—

Math DeepSeek-V3.1 leads

Amazon Nova Pro: 28.5 (#243), DeepSeek-V3.1: 38.9 (#122)

BenchmarkAmazon Nova ProDeepSeek-V3.1
LMArena Math12521420
Omni-MATH24.2%—
LiveBench Math38%—

Knowledge DeepSeek-V3.1 leads

Amazon Nova Pro: 27.4 (#250), DeepSeek-V3.1: 43.7 (#90)

BenchmarkAmazon Nova ProDeepSeek-V3.1
Vectara Hallucination Rate5.1%5.5%
LMArena Expert12111405
Humanity's Last Exam4.4%—
MMLU-Pro67.3%—
Confabulations30.1%—
GPQA (HELM)44.6%—
MMLU82%—

Multimodal Not comparable

Amazon Nova Pro: 25.0 (#126), DeepSeek-V3.1: —

BenchmarkAmazon Nova ProDeepSeek-V3.1
LMArena Vision980—

Multilingual DeepSeek-V3.1 leads

Amazon Nova Pro: 39.7 (#223), DeepSeek-V3.1: 51.6 (#106)

BenchmarkAmazon Nova ProDeepSeek-V3.1
LMArena Non-English12341400
LMArena Chinese12441469
LMArena French12711447
LMArena German12431411
LMArena Japanese12001378
LMArena Korean12031337
LMArena Russian12401405
LMArena Spanish11821431

Instruction Following DeepSeek-V3.1 leads

Amazon Nova Pro: 64.9 (#226), DeepSeek-V3.1: 73.9 (#110)

BenchmarkAmazon Nova ProDeepSeek-V3.1
LMArena Instruction Following12351400
LiveBench Instruction Following67.1%—
IFEval81.5%—

Long Context Amazon Nova Pro leads

Amazon Nova Pro: 38.1 (#205), DeepSeek-V3.1: 36.3 (#232)

BenchmarkAmazon Nova ProDeepSeek-V3.1
LMArena Longer Query12551422
Fiction.LiveBench—52.8%

Writing & Preference DeepSeek-V3.1 leads

Amazon Nova Pro: 43.9 (#226), DeepSeek-V3.1: 60.3 (#98)

BenchmarkAmazon Nova ProDeepSeek-V3.1
LMArena Text12591420
LMArena Creative Writing12121401
LMArena Multi-Turn12461408
Short-Story Creative Writing60.5%—
EQ-Bench Creative Writing—1436
WildBench77.7%—
LiveBench Language37%—

Frequently asked questions

Is Amazon Nova Pro better than DeepSeek-V3.1?

DeepSeek-V3.1 is the stronger model overall, scoring 42.8 to 31.0 on the Noometry Index.

Which is cheaper, Amazon Nova Pro or DeepSeek-V3.1?

DeepSeek-V3.1 is cheaper. It lists at $0.25 per million input tokens and $0.95 per million output tokens; Amazon Nova Pro lists at $0.80 and $3.20.

Is Amazon Nova Pro or DeepSeek-V3.1 better for coding?

DeepSeek-V3.1 scores higher on coding benchmarks: 40.3 versus 35.1 in the Noometry coding category.

Which has the bigger context window?

Amazon Nova Pro does, with 300K tokens against 164K.

How many benchmarks do Amazon Nova Pro and DeepSeek-V3.1 share?

19 benchmarks have published results for both models. Amazon Nova Pro has 38 scored results on Noometry and DeepSeek-V3.1 has 27.

Related comparisons

Go deeper