Noometry

Model comparison

Claude 3 Haiku vs o1

o1 is the stronger model overall, scoring 40.9 to 25.9 on the Noometry Index.

Last verified . 29 shared benchmarks.

Claude 3 Haiku Anthropic

25.9

Rank #340 Confirmed

o1 OpenAI

40.9

Rank #143 Confirmed

Summary

  • They share 29 benchmarks with published results for both. Claude 3 Haiku scores higher in 0 categories and o1 in 9 categories; 9 gaps are clear of the uncertainty.
  • The widest gap is in math, where o1 leads 36.1 to 9.8.
  • The biggest single-benchmark swing is MATH Level 5: 14.9% for Claude 3 Haiku and 94.7% for o1.

Side by side

Claude 3 Haikuo1
ProviderAnthropicOpenAI
Noometry Index25.940.9
Released2024-03-072024-09-12
WeightsProprietaryProprietary
Context window—200K
Max output—100K
Input $ / M tokens—$15
Output $ / M tokens—$60
Results tracked3752

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

Category by category

Coding o1 leads

Claude 3 Haiku: 26.4 (#325), o1: 46.1 (#70)

BenchmarkClaude 3 Haikuo1
WeirdML9.8%47.6%
LMArena Coding11991367
CadEval12%56%
HumanEval+68.9%89%
MBPP+68.8%80.2%
Aider Polyglot—61.7%
BigCodeBench Instruct39.4%—
LiveBench Coding—69.7%
BigCodeBench Complete50.1%—

Agentic & Tool Use Not comparable

Claude 3 Haiku: —, o1: 24.6 (#117)

BenchmarkClaude 3 Haikuo1
Cybench—10%
METR Time Horizons—51.1%

Reasoning o1 leads

Claude 3 Haiku: 16.3 (#307), o1: 27.9 (#111)

BenchmarkClaude 3 Haikuo1
LMArena Hard Prompts11741371
DTBench50.1%74.7%
LMCA8.8%22.3%
Epoch Capabilities Index118.35141.91
SimpleBench—41.7%
Kagi LLM Benchmark34.2%—
ARC-AGI-1—30.7%
Chess Puzzles—15%
EnigmaEval—5.7%
LiveBench Reasoning—91.6%
LiveBench Data Analysis—65.5%
ForecastBench53.2—
LiveBench—75.7%
WinoGrande74.2%—

Math o1 leads

Claude 3 Haiku: 9.8 (#319), o1: 36.1 (#175)

BenchmarkClaude 3 Haikuo1
OTIS Mock AIME 2024-20251.8%73.3%
LMArena Math11881388
MATH Level 514.9%94.7%
FrontierMath (Tiers 1-3)—14.7%
LiveBench Math—80.3%
FrontierMath (Feb 2025 set)—9.3%

Knowledge o1 leads

Claude 3 Haiku: 17.3 (#285), o1: 41.5 (#110)

BenchmarkClaude 3 Haikuo1
GPQA Diamond36.3%76.8%
Confabulations34.2%11.7%
LMArena Expert11481361
Humanity's Last Exam—8%
SimpleQA Verified—41.1%
MMLU73.8%—

Multimodal o1 leads

Claude 3 Haiku: 23.6 (#128), o1: 34.2 (#93)

BenchmarkClaude 3 Haikuo1
LMArena Vision9501168
GeoBench—80%
VPCT—37%
ScienceQA72%—
SpatialViz-Bench—41.4%

Multilingual o1 leads

Claude 3 Haiku: 36.0 (#243), o1: 48.6 (#142)

BenchmarkClaude 3 Haikuo1
LMArena Non-English11781358
LMArena Chinese11551394
LMArena French11951344
LMArena German11741337
LMArena Japanese11021346
LMArena Korean11091396
LMArena Russian12041356
LMArena Spanish11661345

Instruction Following o1 leads

Claude 3 Haiku: 61.3 (#247), o1: 74.8 (#86)

BenchmarkClaude 3 Haikuo1
LMArena Instruction Following11731367
LiveBench Instruction Following—81.5%

Long Context o1 leads

Claude 3 Haiku: 36.1 (#237), o1: 50.3 (#9)

BenchmarkClaude 3 Haikuo1
LMArena Longer Query11901378
Fiction.LiveBench—83.3%

Writing & Preference o1 leads

Claude 3 Haiku: 29.7 (#291), o1: 55.6 (#144)

BenchmarkClaude 3 Haikuo1
LMArena Text11951366
LMArena Creative Writing11571348
LMArena Multi-Turn11901369
Short-Story Creative Writing—70.2%
EQ-Bench Creative Writing717—
LiveBench Language—65.4%

Frequently asked questions

Is Claude 3 Haiku better than o1?

o1 is the stronger model overall, scoring 40.9 to 25.9 on the Noometry Index.

Is Claude 3 Haiku or o1 better for coding?

o1 scores higher on coding benchmarks: 46.1 versus 26.4 in the Noometry coding category.

How many benchmarks do Claude 3 Haiku and o1 share?

29 benchmarks have published results for both models. Claude 3 Haiku has 37 scored results on Noometry and o1 has 52.

Related comparisons

Go deeper