Noometry

Model comparison

Claude 3 Haiku vs Llama 3-8B

Claude 3 Haiku and Llama 3-8B score almost the same on the Noometry Index (25.9 vs 25.5), so choose on price, context window or the category you care about most.

Last verified . 29 shared benchmarks.

Claude 3 Haiku Anthropic

25.9

Rank #340 Confirmed

Llama 3-8B Meta

25.5

Rank #344 Confirmed

Summary

  • They share 29 benchmarks with published results for both. Claude 3 Haiku scores higher in 6 categories and Llama 3-8B in 2 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in knowledge, where Claude 3 Haiku leads 17.3 to 7.8.
  • The biggest single-benchmark swing is BigCodeBench Complete: 50.1% for Claude 3 Haiku and 36.9% for Llama 3-8B.
  • Llama 3-8B has downloadable open weights; the other is API-only.

Side by side

Claude 3 HaikuLlama 3-8B
ProviderAnthropicMeta
Noometry Index25.925.5
Released2024-03-072024-04-18
WeightsProprietaryOpen
Context window——
Max output——
Input $ / M tokens——
Output $ / M tokens——
Results tracked3734

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

Category by category

Coding Llama 3-8B leads

Claude 3 Haiku: 26.4 (#325), Llama 3-8B: 31.0 (#289)

BenchmarkClaude 3 HaikuLlama 3-8B
BigCodeBench Instruct39.4%31.9%
LMArena Coding11991152
BigCodeBench Complete50.1%36.9%
HumanEval+68.9%56.7%
MBPP+68.8%54.8%
WeirdML9.8%—
CadEval12%—

Reasoning Claude 3 Haiku leads

Claude 3 Haiku: 16.3 (#307), Llama 3-8B: 14.3 (#326)

BenchmarkClaude 3 HaikuLlama 3-8B
LMArena Hard Prompts11741133
DTBench50.1%43.9%
Epoch Capabilities Index118.35116.45
ForecastBench53.258.6
WinoGrande74.2%75.7%
Kagi LLM Benchmark34.2%—
Chess Puzzles—0%
LMCA8.8%—
Adversarial NLI—57.3%

Math Claude 3 Haiku leads

Claude 3 Haiku: 9.8 (#319), Llama 3-8B: 8.8 (#323)

BenchmarkClaude 3 HaikuLlama 3-8B
OTIS Mock AIME 2024-20251.8%1.9%
LMArena Math11881151
MATH Level 514.9%6.1%

Knowledge Claude 3 Haiku leads

Claude 3 Haiku: 17.3 (#285), Llama 3-8B: 7.8 (#308)

BenchmarkClaude 3 HaikuLlama 3-8B
GPQA Diamond36.3%26.1%
LMArena Expert11481113
MMLU73.8%68.8%
Confabulations34.2%—
ARC (AI2) Challenge—82.8%
OpenBookQA—82.6%
TriviaQA—67.7%

Multimodal Not comparable

Claude 3 Haiku: 23.6 (#128), Llama 3-8B: —

BenchmarkClaude 3 HaikuLlama 3-8B
LMArena Vision950—
ScienceQA72%—

Multilingual Claude 3 Haiku leads

Claude 3 Haiku: 36.0 (#243), Llama 3-8B: 30.8 (#261)

BenchmarkClaude 3 HaikuLlama 3-8B
LMArena Non-English11781098
LMArena Chinese11551076
LMArena French11951159
LMArena German11741104
LMArena Japanese1102967
LMArena Korean11091004
LMArena Russian12041109
LMArena Spanish11661173

Instruction Following Claude 3 Haiku leads

Claude 3 Haiku: 61.3 (#247), Llama 3-8B: 58.4 (#260)

BenchmarkClaude 3 HaikuLlama 3-8B
LMArena Instruction Following11731127

Long Context Claude 3 Haiku leads

Claude 3 Haiku: 36.1 (#237), Llama 3-8B: 34.2 (#251)

BenchmarkClaude 3 HaikuLlama 3-8B
LMArena Longer Query11901128

Writing & Preference Llama 3-8B leads

Claude 3 Haiku: 29.7 (#291), Llama 3-8B: 37.5 (#256)

BenchmarkClaude 3 HaikuLlama 3-8B
LMArena Text11951166
LMArena Creative Writing11571150
LMArena Multi-Turn11901152
EQ-Bench Creative Writing717—

Frequently asked questions

Is Claude 3 Haiku better than Llama 3-8B?

Claude 3 Haiku and Llama 3-8B score almost the same on the Noometry Index (25.9 vs 25.5), so choose on price, context window or the category you care about most.

Is Claude 3 Haiku or Llama 3-8B better for coding?

Llama 3-8B scores higher on coding benchmarks: 31.0 versus 26.4 in the Noometry coding category.

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

29 benchmarks have published results for both models. Claude 3 Haiku has 37 scored results on Noometry and Llama 3-8B has 34.

Related comparisons

Go deeper