Noometry

Model comparison

Claude 3 Haiku vs Llama 2-70B

Claude 3 Haiku is the stronger model overall, scoring 25.9 to 24.4 on the Noometry Index.

Last verified . 25 shared benchmarks.

Claude 3 Haiku Anthropic

25.9

Rank #340 Confirmed

Llama 2-70B Meta

24.4

Rank #349 Confirmed

Summary

  • They share 25 benchmarks with published results for both. Claude 3 Haiku scores higher in 6 categories and Llama 2-70B in 2 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in knowledge, where Claude 3 Haiku leads 17.3 to 7.4.
  • The biggest single-benchmark swing is MATH Level 5: 14.9% for Claude 3 Haiku and 3.3% for Llama 2-70B.
  • Llama 2-70B has downloadable open weights; the other is API-only.

Side by side

Claude 3 HaikuLlama 2-70B
ProviderAnthropicMeta
Noometry Index25.924.4
Released2024-03-072023-07-18
WeightsProprietaryOpen
Context window——
Max output——
Input $ / M tokens——
Output $ / M tokens——
Results tracked3735

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

Category by category

Coding Llama 2-70B leads

Claude 3 Haiku: 26.4 (#325), Llama 2-70B: 31.4 (#286)

BenchmarkClaude 3 HaikuLlama 2-70B
LMArena Coding11991079
WeirdML9.8%—
BigCodeBench Instruct39.4%—
BigCodeBench Complete50.1%—
CadEval12%—
HumanEval+68.9%—
MBPP+68.8%—

Reasoning Claude 3 Haiku leads

Claude 3 Haiku: 16.3 (#307), Llama 2-70B: 14.4 (#325)

BenchmarkClaude 3 HaikuLlama 2-70B
LMArena Hard Prompts11741073
DTBench50.1%41.6%
Epoch Capabilities Index118.35113.79
ForecastBench53.251.4
WinoGrande74.2%80.2%
Kagi LLM Benchmark34.2%—
LMCA8.8%—
BIG-Bench Hard—64.9%
CommonsenseQA 2.0—50%
HellaSwag—85.3%
LAMBADA—78.9%
PIQA—82.8%

Math Claude 3 Haiku leads

Claude 3 Haiku: 9.8 (#319), Llama 2-70B: 8.1 (#326)

BenchmarkClaude 3 HaikuLlama 2-70B
OTIS Mock AIME 2024-20251.8%0%
LMArena Math11881091
MATH Level 514.9%3.3%
GSM8K—69.6%

Knowledge Claude 3 Haiku leads

Claude 3 Haiku: 17.3 (#285), Llama 2-70B: 7.4 (#310)

BenchmarkClaude 3 HaikuLlama 2-70B
GPQA Diamond36.3%26.3%
LMArena Expert11481039
MMLU73.8%69.9%
Confabulations34.2%—
ARC (AI2) Challenge—78.3%
BoolQ—88.6%
OpenBookQA—60.2%
TriviaQA—87.6%

Multimodal Not comparable

Claude 3 Haiku: 23.6 (#128), Llama 2-70B: —

BenchmarkClaude 3 HaikuLlama 2-70B
LMArena Vision950—
ScienceQA72%—

Multilingual Claude 3 Haiku leads

Claude 3 Haiku: 36.0 (#243), Llama 2-70B: 27.7 (#274)

BenchmarkClaude 3 HaikuLlama 2-70B
LMArena Non-English11781045
LMArena Chinese1155995
LMArena French11951090
LMArena German11741041
LMArena Japanese1102927
LMArena Korean1109964
LMArena Russian12041083
LMArena Spanish11661143

Instruction Following Claude 3 Haiku leads

Claude 3 Haiku: 61.3 (#247), Llama 2-70B: 54.9 (#278)

BenchmarkClaude 3 HaikuLlama 2-70B
LMArena Instruction Following11731071

Long Context Claude 3 Haiku leads

Claude 3 Haiku: 36.1 (#237), Llama 2-70B: 32.3 (#270)

BenchmarkClaude 3 HaikuLlama 2-70B
LMArena Longer Query11901062

Writing & Preference Llama 2-70B leads

Claude 3 Haiku: 29.7 (#291), Llama 2-70B: 32.3 (#279)

BenchmarkClaude 3 HaikuLlama 2-70B
LMArena Text11951115
LMArena Creative Writing11571075
LMArena Multi-Turn11901088
EQ-Bench Creative Writing717—

Frequently asked questions

Is Claude 3 Haiku better than Llama 2-70B?

Claude 3 Haiku is the stronger model overall, scoring 25.9 to 24.4 on the Noometry Index.

Is Claude 3 Haiku or Llama 2-70B better for coding?

Llama 2-70B scores higher on coding benchmarks: 31.4 versus 26.4 in the Noometry coding category.

How many benchmarks do Claude 3 Haiku and Llama 2-70B share?

25 benchmarks have published results for both models. Claude 3 Haiku has 37 scored results on Noometry and Llama 2-70B has 35.

Related comparisons

Go deeper