Claude 3.5 Haiku logo

Claude 3.5 Haiku

Anthropic

Claude 3.5 Haiku is Anthropic's speediest AI model, providing sophisticated coding, tool utilization, and logical deduction skills at a budget-friendly cost. It's perfect for powering user-focused applications, managing specialized sub-agent functions, and creating customized experiences derived from extensive datasets. This model shines in areas like code completion, interactive chatbot interactions, data mining, and immediate content oversight.

Model Specifications

Technical details and capabilities of Claude 3.5 Haiku

Core Specifications

200.0K / 200.0K

Input / Output tokens

October 21, 2024

Release date

Capabilities & License

Multimodal Support
Not Supported
Web Hydrated
No
License
Proprietary

Resources

API Reference
https://docs.anthropic.com/en/docs/intro-to-claude#claude-3-5-family
Playground
https://claude.ai

Performance Insights

Check out how Claude 3.5 Haiku handles various AI tasks through comprehensive benchmark results.

90
68
45
23
0
88.1
HumanEval
88.1
(98%)
85.6
MGSM
85.6
(95%)
83.1
DROP
83.1
(92%)
69.4
MATH
69.4
(77%)
65
MMLU-Pro
65
(72%)
51
TAU-bench Retail
51
(57%)
41.6
GPQA
41.6
(46%)
40.6
SWE-bench Verified
40.6
(45%)
22.8
TAU-bench Airline
22.8
(25%)
HumanEval
MGSM
DROP
MATH
MMLU-Pro
TAU-bench Retail
GPQA
SWE-bench Verified
TAU-bench Airline

Model Comparison

See how Claude 3.5 Haiku stacks up against other leading models across key performance metrics.

100
80
60
40
20
0
41.6
GPQA - Claude 3.5 Haiku
41.6
(42%)
40.2
GPQA - GPT-4o mini
40.2
(40%)
48
GPQA - GPT-4 Turbo
48
(48%)
50.4
GPQA - Claude 3 Opus
50.4
(50%)
53.6
GPQA - GPT-4o
53.6
(54%)
59.4
GPQA - Claude 3.5 Sonnet
59.4
(59%)
88.1
HumanEval - Claude 3.5 Haiku
88.1
(88%)
87.2
HumanEval - GPT-4o mini
87.2
(87%)
87.1
HumanEval - GPT-4 Turbo
87.1
(87%)
84.9
HumanEval - Claude 3 Opus
84.9
(85%)
90.2
HumanEval - GPT-4o
90.2
(90%)
92
HumanEval - Claude 3.5 Sonnet
92
(92%)
69.4
MATH - Claude 3.5 Haiku
69.4
(69%)
70.2
MATH - GPT-4o mini
70.2
(70%)
72.6
MATH - GPT-4 Turbo
72.6
(73%)
60.1
MATH - Claude 3 Opus
60.1
(60%)
76.6
MATH - GPT-4o
76.6
(77%)
71.1
MATH - Claude 3.5 Sonnet
71.1
(71%)
85.6
MGSM - Claude 3.5 Haiku
85.6
(86%)
87
MGSM - GPT-4o mini
87
(87%)
88.5
MGSM - GPT-4 Turbo
88.5
(89%)
90.7
MGSM - Claude 3 Opus
90.7
(91%)
90.5
MGSM - GPT-4o
90.5
(91%)
91.6
MGSM - Claude 3.5 Sonnet
91.6
(92%)
83.1
DROP - Claude 3.5 Haiku
83.1
(83%)
79.7
DROP - GPT-4o mini
79.7
(80%)
86
DROP - GPT-4 Turbo
86
(86%)
83.1
DROP - Claude 3 Opus
83.1
(83%)
83.4
DROP - GPT-4o
83.4
(83%)
87.1
DROP - Claude 3.5 Sonnet
87.1
(87%)
GPQA
HumanEval
MATH
MGSM
DROP
Claude 3.5 Haiku
GPT-4o mini
GPT-4 Turbo
Claude 3 Opus
GPT-4o
Claude 3.5 Sonnet

Detailed Benchmarks

Dive deeper into Claude 3.5 Haiku's performance across specific task categories. Expand each section to see detailed metrics and comparisons.

Coding

SWE-bench Verified

Current model
Other models
Avg (47.6%)

HumanEval

Current model
Other models
Avg (83.3%)

Reasoning

DROP

Current model
Other models
Avg (81.9%)

Knowledge

GPQA

Current model
Other models
Avg (44.7%)

MATH

Current model
Other models
Avg (69.1%)

Non categorized

MGSM

Current model
Other models
Avg (82.2%)

TAU-bench Retail

Current model
Other models
Avg (68.7%)

TAU-bench Airline

Current model
Other models
Avg (45.3%)

Providers Pricing Coming Soon

We're working on gathering comprehensive pricing data from all major providers for Claude 3.5 Haiku. Compare costs across platforms to find the best pricing for your use case.

OpenAI
Anthropic
Google
Mistral AI
Cohere

Share your feedback

Hi, I'm Charlie Palars, the founder of Deepranking.ai. I'm always looking for ways to improve the site and make it more useful for you. You can write me through this form or directly through X at @palarsio.

Your feedback helps us improve our service

Stay Ahead with AI Updates

Get insights on Gemini Pro 2.5, Sonnet 3.7 and more top AI models