Noometry

Model comparison

Nova 2 Lite vs Qwen3.5 27B

Qwen3.5 27B is the stronger model overall, scoring 41.4 to 39.0 on the Noometry Index.

Last verified . 18 shared benchmarks.

Nova 2 Lite Amazon

39.0

Rank #161 Confirmed

Qwen3.5 27B Alibaba (Qwen)

41.4

Rank #127 Confirmed

Summary

  • They share 18 benchmarks with published results for both. Nova 2 Lite scores higher in 2 categories and Qwen3.5 27B in 6 categories; 7 gaps are clear of the uncertainty.
  • The widest gap is in writing & preference, where Qwen3.5 27B leads 59.5 to 54.1.
  • The biggest single-benchmark swing is Vectara Hallucination Rate: 5.1% for Nova 2 Lite and 12.1% for Qwen3.5 27B.
  • Both cost about the same: $0.30 input and $2.50 output per million tokens.
  • Nova 2 Lite accepts more context: 1M tokens versus 262K.
  • Qwen3.5 27B has downloadable open weights; the other is API-only.

Side by side

Nova 2 LiteQwen3.5 27B
ProviderAmazonAlibaba (Qwen)
Noometry Index39.041.4
Released2025-12-012026-02-23
WeightsProprietaryOpen
Context window1M262K
Max output64K66K
Input $ / M tokens$0.30$0.30
Output $ / M tokens$2.50$2.40
Results tracked1928

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

Category by category

Coding Nova 2 Lite leads

Nova 2 Lite: 40.1 (#131), Qwen3.5 27B: 38.6 (#159)

BenchmarkNova 2 LiteQwen3.5 27B
LMArena Coding13861427
LMArena WebDev—1357
WeirdML—39.5%
ALE-Bench—349.45

Agentic & Tool Use Not comparable

Nova 2 Lite: 23.9 (#114), Qwen3.5 27B: —

BenchmarkNova 2 LiteQwen3.5 27B
Berkeley Function Calling Leaderboard27.1%—
Vending-Bench 2—201.98

Reasoning Too close to call

Nova 2 Lite: 26.4 (#120), Qwen3.5 27B: 26.6 (#116)

BenchmarkNova 2 LiteQwen3.5 27B
LMArena Hard Prompts13641414
NYT Connections (extended)—47.9%
Thematic Generalization—45.5%
DTBench—82.4%
LMCA—34%

Math Qwen3.5 27B leads

Nova 2 Lite: 34.8 (#154), Qwen3.5 27B: 36.8 (#103)

BenchmarkNova 2 LiteQwen3.5 27B
LMArena Math13591428
MathArena Final-Answer Competitions—56.7%

Knowledge Nova 2 Lite leads

Nova 2 Lite: 42.5 (#88), Qwen3.5 27B: 37.4 (#149)

BenchmarkNova 2 LiteQwen3.5 27B
Vectara Hallucination Rate5.1%12.1%
LMArena Expert13601428

Multimodal Not comparable

Nova 2 Lite: —, Qwen3.5 27B: 38.7 (#56)

BenchmarkNova 2 LiteQwen3.5 27B
LMArena Vision—1240

Multilingual Qwen3.5 27B leads

Nova 2 Lite: 47.3 (#148), Qwen3.5 27B: 51.0 (#112)

BenchmarkNova 2 LiteQwen3.5 27B
LMArena Non-English13371390
LMArena Chinese13641478
LMArena French13811409
LMArena German13431393
LMArena Japanese12731345
LMArena Korean12841359
LMArena Russian13431389
LMArena Spanish13721408

Instruction Following Qwen3.5 27B leads

Nova 2 Lite: 70.7 (#156), Qwen3.5 27B: 73.7 (#114)

BenchmarkNova 2 LiteQwen3.5 27B
LMArena Instruction Following13351393

Long Context Qwen3.5 27B leads

Nova 2 Lite: 40.6 (#145), Qwen3.5 27B: 43.1 (#103)

BenchmarkNova 2 LiteQwen3.5 27B
LMArena Longer Query13351413

Writing & Preference Qwen3.5 27B leads

Nova 2 Lite: 54.1 (#150), Qwen3.5 27B: 59.5 (#106)

BenchmarkNova 2 LiteQwen3.5 27B
LMArena Text13621409
LMArena Creative Writing12911362
LMArena Multi-Turn13381410

Frequently asked questions

Is Nova 2 Lite better than Qwen3.5 27B?

Qwen3.5 27B is the stronger model overall, scoring 41.4 to 39.0 on the Noometry Index.

Which is cheaper, Nova 2 Lite or Qwen3.5 27B?

Qwen3.5 27B is cheaper. It lists at $0.30 per million input tokens and $2.40 per million output tokens; Nova 2 Lite lists at $0.30 and $2.50.

Is Nova 2 Lite or Qwen3.5 27B better for coding?

Nova 2 Lite scores higher on coding benchmarks: 40.1 versus 38.6 in the Noometry coding category.

Which has the bigger context window?

Nova 2 Lite does, with 1M tokens against 262K.

How many benchmarks do Nova 2 Lite and Qwen3.5 27B share?

18 benchmarks have published results for both models. Nova 2 Lite has 19 scored results on Noometry and Qwen3.5 27B has 28.

Related comparisons

Go deeper