NEWHow AI models draw the world map · Gecko Tests
Beta
Home/Models/Claude 3.5 Haiku
Anthropic logo

Claude 3.5 Haiku

by Anthropic · Released Jan 2024

24.5
avg score
Rank #267
Compare
Better than 14% of all models
Context
N/A
Input $/1M
n/a
Output $/1M
n/a
Type
text
License
Proprietary
Benchmarks
18 tested
Data as of
About

Tested on 18 benchmarks · BenchGecko score 24.5. Top scores: HELM — IFEval (79.2%), HELM — WildBench (76.0%), Lech Mazur Writing (73.5%).

Capabilities
coding
30.2
#164 globally
reasoning
76.0
#25 globally
math
18.4
#234 globally
knowledge
39.2
#208 globally
general
27.8
#135 globally
language
79.2
#60 globally
Benchmark Scores
Compare All
Tested on 18 benchmarks · Ranked across 6 categories
Score Distribution (all 312 models)
0255075100
▲ You are here
codingCompare coding →
CadEval

Computer-aided design evaluation. Tests understanding of CAD concepts, 3D modeling, and engineering design principles.

32.0·
WeirdML

Unusual and adversarial machine learning challenges. Tests robustness of reasoning about edge cases in ML systems.

30.7·
Aider polyglot

Multi-language code editing from Aider. Tests editing ability across Python, JavaScript, TypeScript, Java, C++, Go, Rust, and more.

28.0·
reasoningCompare reasoning →
HELM — WildBench

Stanford HELM WildBench evaluation. Tests reasoning on challenging real-world tasks.

76.0·
mathCompare math →
MATH level 5

Competition-level math from AMC, AIME, and olympiad problems. Level 5 is the hardest tier, requiring creative problem-solving.

46.4·
HELM — Omni-MATH

Stanford HELM evaluation of mathematical reasoning across diverse problem types.

22.4·
OTIS Mock AIME 2024-2025

Mock AIME (American Invitational Mathematics Exam) problems from OTIS. Tests mathematical competition performance.

4.2·
Quick compare:Alibaba Qwen logovs CodeQwen1.5-7BMistral AI logovs Mixtral 8x22B InstructMistral AI logovs Mistral Medium 3.5
Excellent (85+) Good (70-85) Average (50-70) Below (<50)
Model Family · Anthropic Claude Haiku
Mar 2024
26.3
$0.25/M in200Kctx10 benchmarks
Jan 2024
24.5-1.8
n/an/actx18 benchmarks
Oct 2025
35.9+11.4
$1.00/M in200Kctx17 benchmarks
Oct 2025
$0.50/M in(-0.50)200Kctx
See the full Claude Haiku family →
Similar Models
Alibaba Qwen logo
CodeQwen1.5-7B
Alibaba Qwen
24.6n/a
Mistral AI logo
Mixtral 8x22B Instruct
Mistral AI
24.3$2.00/1M
Mistral AI logo
Mistral Medium 3.5
Mistral AI
25.0$1.50/1M
Links
Info
AnthropicPricing explorerDevelopers · API
Research
Technical Report
Documentation
API DocsPlayground
Community
@Anthropic
BenchGecko API
claude-3-5-haiku
Specifications
  • Typetext
  • ContextN/A
  • ReleasedJan 2024
  • LicenseProprietary
  • Statusbenchmark-only
Available On
Anthropic logoAnthropicn/a
Categories
codingreasoningmathknowledgegenerallanguage
Learn More
transformertokens
Share & Export
Tweet
Related Models
Alibaba Qwen logoCodeQwen1.5-7B
24.6
Mistral AI logoMixtral 8x22B Instruct
24.3
Mistral AI logoMistral Medium 3.5
25.0
Alibaba logoQwen2 1.5B Instruct
23.9
Claude 3.5 Haiku is a proprietary text AI model by Anthropic, released in January 2024. It has an average benchmark score of 24.5.

Related Models

CodeQwen1.5-7B · Alibaba QwenMixtral 8x22B Instruct · Mistral AIMistral Medium 3.5 · Mistral AIQwen2 1.5B Instruct · AlibabaMistral Small 3.2 24B · Mistral AI

Benchmarks

HELM — IFEvalHELM — WildBenchLech Mazur WritingMMLUHELM — MMLU-Pro

Related Pages

Anthropic · ProviderAnthropic · EconomyAll ModelsCompare ModelsPricingDevelopers · API

Key facts · as of 2026-05-03

  • Claude 3.5 Haiku by Anthropic. BenchGecko score 24.5, rank 267 of 312 scored models (normalized average of public benchmark scores).
  • List price n/a input · n/a output per 1M tokens (as of 2026-05-03).

How to cite · data as of 2026-05-03

Claude 3.5 Haiku · benchmarks, pricing and providers. BenchGecko, data as of 2026-05-03. https://benchgecko.ai/model/claude-3-5-haiku

Credit "Source: BenchGecko" with a link. Prices per provider and Gecko Tests are BenchGecko data (CC BY 4.0); benchmark scores keep their original source, listed in the JSON. JSON · llms.txt · MCP