Noometry

Model comparison

GPT-4 Turbo vs Nova 2 Lite

Nova 2 Lite is the stronger model overall, scoring 39.7 to 30.5 on the Noometry Index.

Last verified . 17 shared benchmarks.

GPT-4 Turbo OpenAI

30.5

Rank #292 Confirmed

Nova 2 Lite Amazon

39.7

Rank #161 Confirmed

Summary

  • They share 17 benchmarks with published results for both. GPT-4 Turbo scores higher in 0 categories and Nova 2 Lite in 8 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in math, where Nova 2 Lite leads 37.5 to 9.0.
  • Nova 2 Lite is cheaper at $0.30 / $2.50 per million input/output tokens, against $10 / $30 for GPT-4 Turbo.
  • Nova 2 Lite accepts more context: 1M tokens versus 128K.

Side by side

GPT-4 TurboNova 2 Lite
ProviderOpenAIAmazon
Noometry Index30.539.7
Released2023-11-062025-12-01
WeightsProprietaryProprietary
Context window128K1M
Max output4K64K
Input $ / M tokens$10$0.30
Output $ / M tokens$30$2.50
Results tracked3619

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

Category by category

Coding Nova 2 Lite leads

GPT-4 Turbo: 33.8 (#249), Nova 2 Lite: 40.7 (#134)

BenchmarkGPT-4 TurboNova 2 Lite
LMArena Coding12681385
WeirdML18%—
BigCodeBench Instruct48.2%—
BigCodeBench Complete58.2%—
HumanEval+86.6%—
MBPP+73.3%—

Agentic & Tool Use Not comparable

GPT-4 Turbo: —, Nova 2 Lite: 24.1 (#121)

BenchmarkGPT-4 TurboNova 2 Lite
Berkeley Function Calling Leaderboard—27.1%
METR Time Horizons36.7%—

Reasoning Nova 2 Lite leads

GPT-4 Turbo: 15.3 (#317), Nova 2 Lite: 27.5 (#118)

BenchmarkGPT-4 TurboNova 2 Lite
LMArena Hard Prompts12511364
SimpleBench25.1%—
Chess Puzzles6%—
DTBench61.6%—
LMCA9.8%—
Epoch Capabilities Index127.25—
ForecastBench59.4—

Math Nova 2 Lite leads

GPT-4 Turbo: 9.0 (#322), Nova 2 Lite: 37.5 (#156)

BenchmarkGPT-4 TurboNova 2 Lite
LMArena Math12721359
FrontierMath (Tiers 1-3)0.7%—
OTIS Mock AIME 2024-20256.7%—
MATH Level 546.7%—

Knowledge Nova 2 Lite leads

GPT-4 Turbo: 24.3 (#268), Nova 2 Lite: 43.0 (#94)

BenchmarkGPT-4 TurboNova 2 Lite
LMArena Expert12231358
GPQA Diamond46.6%—
Confabulations28.4%—
Vectara Hallucination Rate—5.1%
MMLU81.3%—

Multimodal Not comparable

GPT-4 Turbo: 30.6 (#110), Nova 2 Lite: —

BenchmarkGPT-4 TurboNova 2 Lite
LMArena Vision1090—

Multilingual Nova 2 Lite leads

GPT-4 Turbo: 40.5 (#216), Nova 2 Lite: 47.1 (#153)

BenchmarkGPT-4 TurboNova 2 Lite
LMArena Non-English12451337
LMArena Chinese12421364
LMArena French12761381
LMArena German12591343
LMArena Japanese11941271
LMArena Korean11871284
LMArena Russian12591343
LMArena Spanish12601373

Instruction Following Nova 2 Lite leads

GPT-4 Turbo: 65.8 (#216), Nova 2 Lite: 70.5 (#161)

BenchmarkGPT-4 TurboNova 2 Lite
LMArena Instruction Following12491335

Long Context Nova 2 Lite leads

GPT-4 Turbo: 38.0 (#206), Nova 2 Lite: 40.6 (#150)

BenchmarkGPT-4 TurboNova 2 Lite
LMArena Longer Query12541335

Writing & Preference Nova 2 Lite leads

GPT-4 Turbo: 47.7 (#206), Nova 2 Lite: 53.9 (#154)

BenchmarkGPT-4 TurboNova 2 Lite
LMArena Text12721362
LMArena Creative Writing12691291
LMArena Multi-Turn12671338

Frequently asked questions

Is GPT-4 Turbo better than Nova 2 Lite?

Nova 2 Lite is the stronger model overall, scoring 39.7 to 30.5 on the Noometry Index.

Which is cheaper, GPT-4 Turbo or Nova 2 Lite?

Nova 2 Lite is cheaper. It lists at $0.30 per million input tokens and $2.50 per million output tokens; GPT-4 Turbo lists at $10 and $30.

Is GPT-4 Turbo or Nova 2 Lite better for coding?

Nova 2 Lite scores higher on coding benchmarks: 40.7 versus 33.8 in the Noometry coding category.

Which has the bigger context window?

Nova 2 Lite does, with 1M tokens against 128K.

How many benchmarks do GPT-4 Turbo and Nova 2 Lite share?

17 benchmarks have published results for both models. GPT-4 Turbo has 36 scored results on Noometry and Nova 2 Lite has 19.

Related comparisons

Go deeper