Model comparison

Amazon Nova Pro vs Hy3

Hy3 is the stronger model overall, scoring 43.6 to 30.5 on the Noometry Index.

Last verified . 15 shared benchmarks.

Amazon Nova Pro Amazon

30.5

Rank #278 Confirmed

Hy3 Tencent

43.6

Rank #79 Confirmed

Summary

  • They share 15 benchmarks with published results for both. Amazon Nova Pro scores higher in 0 categories and Hy3 in 8 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in writing & preference, where Hy3 leads 62.5 to 44.1.
  • Hy3 is cheaper at $0.13 / $0.53 per million input/output tokens, against $0.80 / $3.20 for Amazon Nova Pro.
  • Amazon Nova Pro accepts more context: 300K tokens versus 262K.
  • Hy3 has downloadable open weights; the other is API-only.

Side by side

Amazon Nova Pro and Hy3 specifications
Amazon Nova ProHy3
ProviderAmazonTencent
Noometry Index30.543.6
Released2024-12-032026-07-06
WeightsProprietaryOpen
Context window300K262K
Max output10K128K
Input $ / M tokens$0.80$0.13
Output $ / M tokens$3.20$0.53
Results tracked3817

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

Category by category

Coding Hy3 leads

Amazon Nova Pro: 34.6 (#224), Hy3: 46.6 (#60)

Coding benchmarks
BenchmarkAmazon Nova ProHy3
LMArena Coding12701465
LMArena WebDev—1509
LiveBench Coding38.1%—

Agentic & Tool Use Not comparable

Amazon Nova Pro: 16.6 (#139), Hy3: —

Agentic & Tool Use benchmarks
BenchmarkAmazon Nova ProHy3
Berkeley Function Calling Leaderboard25%—
TheAgentCompany1.7%—

Reasoning Hy3 leads

Amazon Nova Pro: 19.2 (#240), Hy3: 25.0 (#134)

Reasoning benchmarks
BenchmarkAmazon Nova ProHy3
LMArena Hard Prompts12461449
NYT Connections (extended)—41.2%
LiveBench Reasoning32.6%—
LiveBench Data Analysis48.3%—
Epoch Capabilities Index123.75—
LiveBench43.5%—

Math Hy3 leads

Amazon Nova Pro: 26.4 (#240), Hy3: 37.4 (#98)

Math benchmarks
BenchmarkAmazon Nova ProHy3
LMArena Math12521475
Omni-MATH24.2%—
LiveBench Math38%—

Knowledge Hy3 leads

Amazon Nova Pro: 27.1 (#246), Hy3: 40.2 (#108)

Knowledge benchmarks
BenchmarkAmazon Nova ProHy3
LMArena Expert12111461
Humanity's Last Exam4.4%—
MMLU-Pro67.3%—
Confabulations30.1%—
Vectara Hallucination Rate5.1%—
GPQA (HELM)44.6%—
MMLU82%—

Multimodal Not comparable

Amazon Nova Pro: 24.5 (#121), Hy3: —

Multimodal benchmarks
BenchmarkAmazon Nova ProHy3
LMArena Vision980—

Multilingual Hy3 leads

Amazon Nova Pro: 39.9 (#218), Hy3: 53.7 (#60)

Multilingual benchmarks
BenchmarkAmazon Nova ProHy3
LMArena Non-English12341427
LMArena Chinese12441496
LMArena French12701461
LMArena Korean12031390
LMArena Russian12401435
LMArena Spanish11821448
LMArena German1243—
LMArena Japanese1200—

Instruction Following Hy3 leads

Amazon Nova Pro: 65.1 (#221), Hy3: 75.3 (#68)

Instruction Following benchmarks
BenchmarkAmazon Nova ProHy3
LMArena Instruction Following12351425
LiveBench Instruction Following67.1%—
IFEval81.5%—

Long Context Hy3 leads

Amazon Nova Pro: 38.1 (#199), Hy3: 44.1 (#70)

Long Context benchmarks
BenchmarkAmazon Nova ProHy3
LMArena Longer Query12551443

Writing & Preference Hy3 leads

Amazon Nova Pro: 44.1 (#221), Hy3: 62.5 (#75)

Writing & Preference benchmarks
BenchmarkAmazon Nova ProHy3
LMArena Text12591442
LMArena Creative Writing12121399
LMArena Multi-Turn12451436
Short-Story Creative Writing60.5%—
WildBench77.7%—
LiveBench Language37%—

Frequently asked questions

Is Amazon Nova Pro better than Hy3?

Hy3 is the stronger model overall, scoring 43.6 to 30.5 on the Noometry Index.

Which is cheaper, Amazon Nova Pro or Hy3?

Hy3 is cheaper. It lists at $0.13 per million input tokens and $0.53 per million output tokens; Amazon Nova Pro lists at $0.80 and $3.20.

Is Amazon Nova Pro or Hy3 better for coding?

Hy3 scores higher on coding benchmarks: 46.6 versus 34.6 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 Hy3 share?

15 benchmarks have published results for both models. Amazon Nova Pro has 38 scored results on Noometry and Hy3 has 17.

Related comparisons

Go deeper