Magica logo

Claude Haiku 4.5 vs Claude Sonnet 4.6 (Comparative Analysis)

Loading comparison form...
Comparative Analysis: Claude Haiku 4.5 vs. Claude Sonnet 4.6
Want to try out these models side by side?Try Magica for free

Overview

Claude Haiku 4.5 was released 4 months before Claude Sonnet 4.6.
Claude Haiku 4.5Claude Haiku 4.5
Claude Sonnet 4.6Claude Sonnet 4.6
Model Provider
The organization behind this AI's development
Anthropic logoAnthropic
Anthropic logoAnthropic
Input Context Window
Maximum input tokens this model can process at once
200K
tokens
1M
tokens
Output Token Limit
Maximum output tokens this model can generate at once
64K
tokens
128K
tokens
Release Date
When this model first became publicly available
October 15th, 2025
February 17th, 2026

Capabilities & Features

Compare supported features, modalities, and advanced capabilities
Claude Haiku 4.5
Claude Sonnet 4.6
Input Types
Supported input formats
📝Text🖼️Image📁File
📝Text🖼️Image📁File
Reasoning Controls
Defaults and configurable reasoning effort
Optional
OptionalEffort: Max, High, Medium, LowDefault: Medium
Output Types
Supported output formats
📝Text
📝Text
Tokenizer
Text encoding system
ClaudeClaude
Key Features
Advanced capabilities
Function CallingStructured OutputReasoning ModeContent Moderation
Function CallingStructured OutputReasoning ModeContent Moderation

Pricing

Claude Haiku 4.5 is roughly 0.3x less expensive compared to Claude Sonnet 4.6 for input tokens and roughly 0.3x less expensive for output tokens.
Claude Haiku 4.5Claude Haiku 4.5
Claude Sonnet 4.6Claude Sonnet 4.6
Input Token Cost
Cost per million input tokens
$1.00
per million tokens
$3.00
per million tokens
Output Token Cost
Cost per million output tokens
$5.00
per million tokens
$15.00
per million tokens
Web Search Cost
Additional cost for each web search operation
$0.01
per search
$0.01
per search
Cache Read Cost
Cost to reuse cached input tokens
$0.10
per million tokens
$0.30
per million tokens
Cache Write Cost
Cost to write input tokens to the cache
$1.25
per million tokens
$3.75
per million tokens
1-Hour Cache Write Cost
Cost to cache input tokens for one hour
$2.00
per million tokens
$6.00
per million tokens

Benchmarks

Compare relevant benchmarks between Claude Haiku 4.5 and Claude Sonnet 4.6.
Claude Haiku 4.5Claude Haiku 4.5
Claude Sonnet 4.6Claude Sonnet 4.6
Intelligence Index
Overall model quality across independent evaluations
29.6
(Artificial Analysis index; higher is better)
47.2
(Artificial Analysis index; higher is better)
Coding Index
Programming performance across independent evaluations
43.9
(Artificial Analysis index; higher is better)
63
(Artificial Analysis index; higher is better)
Agentic Index
Ability to complete multi-step agentic tasks
16.4
(Artificial Analysis index; higher is better)
40.8
(Artificial Analysis index; higher is better)
Best Design Arena Score
Highest human-preference Elo score across design arenas
1,177 Elo
(#35 in Asciiart (49.3% win rate))
1,314 Elo
(#14 in Game Dev (60.2% win rate))

At a Glance

Quick overview of what makes Claude Haiku 4.5 and Claude Sonnet 4.6 unique.
Anthropic logoClaude Haiku 4.5 by Anthropic understands both text and images, can use external tools and APIs, offers advanced reasoning, generates structured data. It can handle standard conversations with its 200K token context window. Reasonably priced at $1.00/M input and $5.00/M output tokens. Includes built-in content moderation for safer outputs. Released October 15th, 2025.
Anthropic logoClaude Sonnet 4.6 by Anthropic understands both text and images, can use external tools and APIs, offers advanced reasoning, generates structured data. It can handle standard conversations with its 1M token context window. Premium pricing at $3.00/M input and $15.00/M output tokens. Includes built-in content moderation for safer outputs. Released February 17th, 2026.

Explore More Comparisons

Compare your models with top performers across different categories