Compare AI Models

Select two models to compare their pricing, performance benchmarks, capabilities, and technical specifications side by side.

Model 1

Model 2

Anthropic: Claude Haiku 4.5

anthropic

Select a model

Input Price

Cost per 1M input tokens

$1.00/M tokens

—

Output Price

Cost per 1M output tokens

$5.00/M tokens

—

Cache Read Price

Cost per 1M cached tokens

$0.10/M tokens

—

Cache Write Price

Cost to write 1M tokens to cache

$1.25/M tokens

—

Context Window

Maximum tokens the model can process

200,000 tokens

—

Max Output Tokens

Maximum tokens in model response

64,000 tokens

—

Input Modalities

Supported input types

text, image, file

—

Output Modalities

Supported output types

text

—

LiveCodeBench

Real-world coding task performance

51.1%

—

SciCode

Scientific computing & algorithms

34.4%

—

TerminalBench Hard

CLI & terminal command generation

27.3%

—

Long Context Reasoning

Understanding large codebases

43.7%

—

IFBench

Instruction following accuracy

42.0%

—

Output Speed

Median tokens per second

110.7 tok/s

—

Function Calling

Can call external functions/APIs

✓ Supported

—

Tool Choice

Control over function selection

✓ Supported

—

Structured Outputs

JSON schema validation

✓ Supported

—

Reasoning Tokens

Extended thinking for complex problems

✗ Not supported

—

Content Moderation

Content filtering status

Enabled

—

Model ID

anthropic/claude-haiku-4.5

—

Tokenizer

Claude

—

Release Date

Oct 15, 2025

—

Created On OpenRouter

Oct 15, 2025

—

Compare AI Models

Pricing

Context & Modalities

Performance Benchmarks

Capabilities & Features

Technical Details