Noometry

Model comparison

Gemma 2 9B vs Llama 3.1-405B

Llama 3.1-405B is the stronger model overall, scoring 30.7 to 25.9 on the Noometry Index.

Last verified . 24 shared benchmarks.

Gemma 2 9B Google

25.9

Rank #341 Confirmed

Llama 3.1-405B Meta

30.7

Rank #288 Confirmed

Summary

  • They share 24 benchmarks with published results for both. Gemma 2 9B scores higher in 0 categories and Llama 3.1-405B in 8 categories; 7 gaps are clear of the uncertainty.
  • The widest gap is in knowledge, where Llama 3.1-405B leads 30.4 to 9.7.
  • The biggest single-benchmark swing is MATH Level 5: 21% for Gemma 2 9B and 49.8% for Llama 3.1-405B.

Side by side

Gemma 2 9BLlama 3.1-405B
ProviderGoogleMeta
Noometry Index25.930.7
Released2024-06-242024-07-23
WeightsOpenOpen
Context window——
Max output——
Input $ / M tokens——
Output $ / M tokens——
Results tracked3542

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

Category by category

Coding Llama 3.1-405B leads

Gemma 2 9B: 29.4 (#304), Llama 3.1-405B: 33.1 (#262)

BenchmarkGemma 2 9BLlama 3.1-405B
LMArena Coding11731291
WeirdML—21.4%
BigCodeBench Instruct34.7%—
LiveBench Coding22.5%—
BigCodeBench Complete40.6%—

Agentic & Tool Use Not comparable

Gemma 2 9B: —, Llama 3.1-405B: 21.0 (#140)

BenchmarkGemma 2 9BLlama 3.1-405B
TheAgentCompany—7.4%
Cybench—7.5%

Reasoning Too close to call

Gemma 2 9B: 15.9 (#309), Llama 3.1-405B: 16.8 (#300)

BenchmarkGemma 2 9BLlama 3.1-405B
LMArena Hard Prompts11711269
Epoch Capabilities Index119.83128.75
PIQA83.7%85.9%
SimpleBench—23%
Kagi LLM Benchmark—45%
LiveBench Reasoning15.2%—
DTBench—61.4%
LiveBench Data Analysis36.4%—
BIG-Bench Hard—82.9%
ForecastBench—59.9
HellaSwag—89.2%
LiveBench28.7%—
WinoGrande—89.2%

Math Llama 3.1-405B leads

Gemma 2 9B: 9.9 (#318), Llama 3.1-405B: 18.4 (#290)

BenchmarkGemma 2 9BLlama 3.1-405B
OTIS Mock AIME 2024-20250.6%9.7%
LMArena Math11831281
MATH Level 521%49.8%
Omni-MATH—24.9%
LiveBench Math19.8%—
GSM8K84.9%—

Knowledge Llama 3.1-405B leads

Gemma 2 9B: 9.7 (#305), Llama 3.1-405B: 30.4 (#227)

BenchmarkGemma 2 9BLlama 3.1-405B
GPQA Diamond27.5%50.9%
LMArena Expert11471243
MMLU72.1%84.5%
MMLU-Pro—72.3%
Confabulations—17.6%
GPQA (HELM)—52.2%
ARC (AI2) Challenge—95.3%
BoolQ85.7%—
TriviaQA—82.7%

Multilingual Llama 3.1-405B leads

Gemma 2 9B: 36.6 (#238), Llama 3.1-405B: 40.7 (#214)

BenchmarkGemma 2 9BLlama 3.1-405B
LMArena Non-English11881248
LMArena Chinese11851242
LMArena French11901279
LMArena German11861252
LMArena Japanese11441208
LMArena Korean11371184
LMArena Russian12001265
LMArena Spanish12001260

Instruction Following Llama 3.1-405B leads

Gemma 2 9B: 57.6 (#269), Llama 3.1-405B: 65.9 (#214)

BenchmarkGemma 2 9BLlama 3.1-405B
LMArena Instruction Following11781259
LiveBench Instruction Following52.6%—
IFEval—81.1%

Long Context Llama 3.1-405B leads

Gemma 2 9B: 36.3 (#233), Llama 3.1-405B: 38.4 (#197)

BenchmarkGemma 2 9BLlama 3.1-405B
LMArena Longer Query11971266

Writing & Preference Llama 3.1-405B leads

Gemma 2 9B: 32.1 (#281), Llama 3.1-405B: 38.9 (#251)

BenchmarkGemma 2 9BLlama 3.1-405B
LMArena Text12071284
LMArena Creative Writing12061262
EQ-Bench Creative Writing841870
LMArena Multi-Turn11931297
WildBench—78.3%
LiveBench Language25.5%—

Frequently asked questions

Is Gemma 2 9B better than Llama 3.1-405B?

Llama 3.1-405B is the stronger model overall, scoring 30.7 to 25.9 on the Noometry Index.

Is Gemma 2 9B or Llama 3.1-405B better for coding?

Llama 3.1-405B scores higher on coding benchmarks: 33.1 versus 29.4 in the Noometry coding category.

How many benchmarks do Gemma 2 9B and Llama 3.1-405B share?

24 benchmarks have published results for both models. Gemma 2 9B has 35 scored results on Noometry and Llama 3.1-405B has 42.

Related comparisons

Go deeper