Noometry

Model comparison

Claude 3 Haiku vs StarCoder 2 3B

Claude 3 Haiku has enough public results to be ranked (#340); StarCoder 2 3B does not yet, so treat this comparison as directional.

Last verified . 5 shared benchmarks.

Claude 3 Haiku Anthropic

25.9

Rank #340 Confirmed

StarCoder 2 3B NVIDIA

36.3

Unranked Sparse

Summary

  • They share 5 benchmarks with published results for both. Claude 3 Haiku scores higher in 0 categories and StarCoder 2 3B in 1 category; one gap is clear of the uncertainty.
  • The widest gap is in coding, where StarCoder 2 3B leads 33.5 to 26.4.
  • The biggest single-benchmark swing is BigCodeBench Complete: 50.1% for Claude 3 Haiku and 21.4% for StarCoder 2 3B.
  • StarCoder 2 3B has downloadable open weights; the other is API-only.

Side by side

Claude 3 HaikuStarCoder 2 3B
ProviderAnthropicNVIDIA
Noometry Index25.936.3
Released2024-03-072024-02-22
WeightsProprietaryOpen
Context window——
Max output——
Input $ / M tokens——
Output $ / M tokens——
Results tracked377

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

Category by category

Coding StarCoder 2 3B leads

Claude 3 Haiku: 26.4 (#325), StarCoder 2 3B: 33.5

BenchmarkClaude 3 HaikuStarCoder 2 3B
BigCodeBench Complete50.1%21.4%
HumanEval+68.9%27.4%
WeirdML9.8%—
BigCodeBench Instruct39.4%—
LMArena Coding1199—
CadEval12%—
MBPP+68.8%—

Reasoning Not comparable

Claude 3 Haiku: 16.3 (#307), StarCoder 2 3B: —

BenchmarkClaude 3 HaikuStarCoder 2 3B
Epoch Capabilities Index118.3588.76
WinoGrande74.2%57.1%
Kagi LLM Benchmark34.2%—
LMArena Hard Prompts1174—
DTBench50.1%—
LMCA8.8%—
ForecastBench53.2—

Math Not comparable

Claude 3 Haiku: 9.8 (#319), StarCoder 2 3B: —

BenchmarkClaude 3 HaikuStarCoder 2 3B
OTIS Mock AIME 2024-20251.8%—
LMArena Math1188—
MATH Level 514.9%—
GSM8K—21.6%

Knowledge Not comparable

Claude 3 Haiku: 17.3 (#285), StarCoder 2 3B: —

BenchmarkClaude 3 HaikuStarCoder 2 3B
MMLU73.8%36.6%
GPQA Diamond36.3%—
Confabulations34.2%—
LMArena Expert1148—
ARC (AI2) Challenge—34.2%

Multimodal Not comparable

Claude 3 Haiku: 23.6 (#128), StarCoder 2 3B: —

BenchmarkClaude 3 HaikuStarCoder 2 3B
LMArena Vision950—
ScienceQA72%—

Multilingual Not comparable

Claude 3 Haiku: 36.0 (#243), StarCoder 2 3B: —

BenchmarkClaude 3 HaikuStarCoder 2 3B
LMArena Non-English1178—
LMArena Chinese1155—
LMArena French1195—
LMArena German1174—
LMArena Japanese1102—
LMArena Korean1109—
LMArena Russian1204—
LMArena Spanish1166—

Instruction Following Not comparable

Claude 3 Haiku: 61.3 (#247), StarCoder 2 3B: —

BenchmarkClaude 3 HaikuStarCoder 2 3B
LMArena Instruction Following1173—

Long Context Not comparable

Claude 3 Haiku: 36.1 (#237), StarCoder 2 3B: —

BenchmarkClaude 3 HaikuStarCoder 2 3B
LMArena Longer Query1190—

Writing & Preference Not comparable

Claude 3 Haiku: 29.7 (#291), StarCoder 2 3B: —

BenchmarkClaude 3 HaikuStarCoder 2 3B
LMArena Text1195—
LMArena Creative Writing1157—
EQ-Bench Creative Writing717—
LMArena Multi-Turn1190—

Frequently asked questions

Is Claude 3 Haiku better than StarCoder 2 3B?

Claude 3 Haiku has enough public results to be ranked (#340); StarCoder 2 3B does not yet, so treat this comparison as directional.

Is Claude 3 Haiku or StarCoder 2 3B better for coding?

StarCoder 2 3B scores higher on coding benchmarks: 33.5 versus 26.4 in the Noometry coding category.

How many benchmarks do Claude 3 Haiku and StarCoder 2 3B share?

5 benchmarks have published results for both models. Claude 3 Haiku has 37 scored results on Noometry and StarCoder 2 3B has 7.

Related comparisons

Go deeper