Noometry

Model comparison

Llama 3.2 3B vs Llama 3-8B

Llama 3.2 3B is the stronger model overall, scoring 28.4 to 25.3 on the Noometry Index.

Last verified . 15 shared benchmarks.

Llama 3.2 3B Meta

28.4

Rank #318 Confirmed

Llama 3-8B Meta

25.3

Rank #338 Confirmed

Summary

  • They share 15 benchmarks with published results for both. Llama 3.2 3B scores higher in 3 categories and Llama 3-8B in 5 categories; 7 gaps are clear of the uncertainty.
  • The widest gap is in math, where Llama 3.2 3B leads 29.9 to 8.3.
  • The biggest single-benchmark swing is BigCodeBench Complete: 28.3% for Llama 3.2 3B and 36.9% for Llama 3-8B.
  • Llama 3.2 3B is cheaper at $0.05 / $0.33 per million input/output tokens, against $0.14 / $0.14 for Llama 3-8B.
  • Llama 3.2 3B accepts more context: 131K tokens versus 8K.

Side by side

Llama 3.2 3BLlama 3-8B
ProviderMetaMeta
Noometry Index28.425.3
Released2024-09-242024-04-18
WeightsOpenOpen
Context window131K8K
Max output118K8K
Input $ / M tokens$0.05$0.14
Output $ / M tokens$0.33$0.14
Results tracked1834

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

Category by category

Coding Llama 3-8B leads

Llama 3.2 3B: 27.1 (#314), Llama 3-8B: 30.5 (#285)

BenchmarkLlama 3.2 3BLlama 3-8B
BigCodeBench Instruct23.4%31.9%
LMArena Coding10981152
BigCodeBench Complete28.3%36.9%
HumanEval+—56.7%
MBPP+—54.8%

Agentic & Tool Use Not comparable

Llama 3.2 3B: 19.9 (#135), Llama 3-8B: —

BenchmarkLlama 3.2 3BLlama 3-8B
Berkeley Function Calling Leaderboard21.9%—
BALROG10.1%—

Reasoning Llama 3.2 3B leads

Llama 3.2 3B: 20.1 (#225), Llama 3-8B: 13.8 (#321)

BenchmarkLlama 3.2 3BLlama 3-8B
LMArena Hard Prompts10951133
Chess Puzzles—0%
DTBench—43.9%
Adversarial NLI—57.3%
Epoch Capabilities Index—116.33
ForecastBench—58.6
WinoGrande—75.7%

Math Llama 3.2 3B leads

Llama 3.2 3B: 29.9 (#213), Llama 3-8B: 8.3 (#317)

BenchmarkLlama 3.2 3BLlama 3-8B
LMArena Math11261151
OTIS Mock AIME 2024-2025—1.9%
MATH Level 5—6.1%

Knowledge Llama 3.2 3B leads

Llama 3.2 3B: 29.1 (#232), Llama 3-8B: 7.8 (#303)

BenchmarkLlama 3.2 3BLlama 3-8B
LMArena Expert10901113
GPQA Diamond—26.1%
ARC (AI2) Challenge—82.8%
MMLU—68.8%
OpenBookQA—82.6%
TriviaQA—67.7%

Multilingual Llama 3-8B leads

Llama 3.2 3B: 26.4 (#276), Llama 3-8B: 31.0 (#256)

BenchmarkLlama 3.2 3BLlama 3-8B
LMArena Non-English10191098
LMArena Chinese10171076
LMArena German10561104
LMArena Russian9491109
LMArena French—1158
LMArena Japanese—967
LMArena Korean—1003
LMArena Spanish—1172

Instruction Following Llama 3-8B leads

Llama 3.2 3B: 56.3 (#270), Llama 3-8B: 58.7 (#255)

BenchmarkLlama 3.2 3BLlama 3-8B
LMArena Instruction Following10891127

Long Context Too close to call

Llama 3.2 3B: 33.4 (#256), Llama 3-8B: 34.2 (#246)

BenchmarkLlama 3.2 3BLlama 3-8B
LMArena Longer Query11011128

Writing & Preference Llama 3-8B leads

Llama 3.2 3B: 24.9 (#302), Llama 3-8B: 37.7 (#251)

BenchmarkLlama 3.2 3BLlama 3-8B
LMArena Text11101166
LMArena Creative Writing10941150
LMArena Multi-Turn11051152
EQ-Bench Creative Writing595—

Frequently asked questions

Is Llama 3.2 3B better than Llama 3-8B?

Llama 3.2 3B is the stronger model overall, scoring 28.4 to 25.3 on the Noometry Index.

Which is cheaper, Llama 3.2 3B or Llama 3-8B?

Llama 3.2 3B is cheaper. It lists at $0.05 per million input tokens and $0.33 per million output tokens; Llama 3-8B lists at $0.14 and $0.14.

Is Llama 3.2 3B or Llama 3-8B better for coding?

Llama 3-8B scores higher on coding benchmarks: 30.5 versus 27.1 in the Noometry coding category.

Which has the bigger context window?

Llama 3.2 3B does, with 131K tokens against 8K.

How many benchmarks do Llama 3.2 3B and Llama 3-8B share?

15 benchmarks have published results for both models. Llama 3.2 3B has 18 scored results on Noometry and Llama 3-8B has 34.

Related comparisons

Go deeper