Model comparison

C4ai Aya Expanse 8b vs Nvidia Llama 3.3 Nemotron Super 49b v1.5

Nvidia Llama 3.3 Nemotron Super 49b v1.5 is the stronger model overall, scoring 40.3 to 34.9 on the Noometry Index.

Last verified . 12 shared benchmarks.

C4ai Aya Expanse 8b Cohere

34.9

Rank #229 Confirmed

Summary

  • They share 12 benchmarks with published results for both. C4ai Aya Expanse 8b scores higher in 0 categories and Nvidia Llama 3.3 Nemotron Super 49b v1.5 in 8 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in writing & preference, where Nvidia Llama 3.3 Nemotron Super 49b v1.5 leads 53.1 to 39.0.

Side by side

C4ai Aya Expanse 8b and Nvidia Llama 3.3 Nemotron Super 49b v1.5 specifications
C4ai Aya Expanse 8bNvidia Llama 3.3 Nemotron Super 49b v1.5
ProviderCohereNVIDIA
Noometry Index34.940.3
Released—2025-07-25
WeightsOpenOpen
Context window—131K
Max output—131K
Input $ / M tokens—$0.40
Output $ / M tokens—$0.40
Results tracked1512

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

Category by category

Coding Nvidia Llama 3.3 Nemotron Super 49b v1.5 leads

C4ai Aya Expanse 8b: 33.8 (#252), Nvidia Llama 3.3 Nemotron Super 49b v1.5: 39.8 (#154)

Coding benchmarks
BenchmarkC4ai Aya Expanse 8bNvidia Llama 3.3 Nemotron Super 49b v1.5
LMArena Coding11601355

Reasoning Nvidia Llama 3.3 Nemotron Super 49b v1.5 leads

C4ai Aya Expanse 8b: 22.4 (#197), Nvidia Llama 3.3 Nemotron Super 49b v1.5: 26.8 (#128)

Reasoning benchmarks
BenchmarkC4ai Aya Expanse 8bNvidia Llama 3.3 Nemotron Super 49b v1.5
LMArena Hard Prompts11561336

Math Nvidia Llama 3.3 Nemotron Super 49b v1.5 leads

C4ai Aya Expanse 8b: 33.3 (#203), Nvidia Llama 3.3 Nemotron Super 49b v1.5: 38.2 (#141)

Math benchmarks
BenchmarkC4ai Aya Expanse 8bNvidia Llama 3.3 Nemotron Super 49b v1.5
LMArena Math11681392

Knowledge Nvidia Llama 3.3 Nemotron Super 49b v1.5 leads

C4ai Aya Expanse 8b: 33.6 (#201), Nvidia Llama 3.3 Nemotron Super 49b v1.5: 36.7 (#165)

Knowledge benchmarks
BenchmarkC4ai Aya Expanse 8bNvidia Llama 3.3 Nemotron Super 49b v1.5
LMArena Expert11531330
Vectara Hallucination Rate9.5%—

Multilingual Nvidia Llama 3.3 Nemotron Super 49b v1.5 leads

C4ai Aya Expanse 8b: 36.1 (#242), Nvidia Llama 3.3 Nemotron Super 49b v1.5: 45.5 (#168)

Multilingual benchmarks
BenchmarkC4ai Aya Expanse 8bNvidia Llama 3.3 Nemotron Super 49b v1.5
LMArena Non-English11801316
LMArena Japanese11201300
LMArena Russian11971332
LMArena Chinese1181—
LMArena German1188—

Instruction Following Nvidia Llama 3.3 Nemotron Super 49b v1.5 leads

C4ai Aya Expanse 8b: 60.1 (#253), Nvidia Llama 3.3 Nemotron Super 49b v1.5: 68.6 (#188)

Instruction Following benchmarks
BenchmarkC4ai Aya Expanse 8bNvidia Llama 3.3 Nemotron Super 49b v1.5
LMArena Instruction Following11551299

Long Context Nvidia Llama 3.3 Nemotron Super 49b v1.5 leads

C4ai Aya Expanse 8b: 36.1 (#236), Nvidia Llama 3.3 Nemotron Super 49b v1.5: 40.0 (#164)

Long Context benchmarks
BenchmarkC4ai Aya Expanse 8bNvidia Llama 3.3 Nemotron Super 49b v1.5
LMArena Longer Query11911315

Writing & Preference Nvidia Llama 3.3 Nemotron Super 49b v1.5 leads

C4ai Aya Expanse 8b: 39.0 (#250), Nvidia Llama 3.3 Nemotron Super 49b v1.5: 53.1 (#159)

Writing & Preference benchmarks
BenchmarkC4ai Aya Expanse 8bNvidia Llama 3.3 Nemotron Super 49b v1.5
LMArena Text11851338
LMArena Creative Writing11681307
LMArena Multi-Turn11601334

Frequently asked questions

Is C4ai Aya Expanse 8b better than Nvidia Llama 3.3 Nemotron Super 49b v1.5?

Nvidia Llama 3.3 Nemotron Super 49b v1.5 is the stronger model overall, scoring 40.3 to 34.9 on the Noometry Index.

Is C4ai Aya Expanse 8b or Nvidia Llama 3.3 Nemotron Super 49b v1.5 better for coding?

Nvidia Llama 3.3 Nemotron Super 49b v1.5 scores higher on coding benchmarks: 39.8 versus 33.8 in the Noometry coding category.

How many benchmarks do C4ai Aya Expanse 8b and Nvidia Llama 3.3 Nemotron Super 49b v1.5 share?

12 benchmarks have published results for both models. C4ai Aya Expanse 8b has 15 scored results on Noometry and Nvidia Llama 3.3 Nemotron Super 49b v1.5 has 12.

Related comparisons

Go deeper