anthropic

Anthropic: Claude Haiku 4.5

The Anthropic: Claude Haiku 4.5 is an AI model that can process text, images, and files as input modalities, with a context length of 200,000 tokens. It also supports tools and reasoning capabilities. This model may be worth considering for those who need to handle diverse input types and have a budget for its pricing structure: $1 per million input tokens and $5 per million output tokens. Its blended benchmark score of 29.4 across one independent benchmark is a point to consider, but further evaluation would be necessary to determine its overall performance in specific use cases.

Quality Score
100/100
price + capability + benchmarks
Input Price
$1.00
per 1M tokens · checked 2026-09-07
Output Price
$5.00
per 1M tokens · checked 2026-09-07
Context Window
200,000
tokens

Benchmark results

Independent, published benchmarks. Blended score 29.4 across 1 benchmark, last refreshed 2026-09-07. How scoring works →

BenchmarkMeasuresScore
AI Index broad capability composite 28.7
Model ID
anthropic/claude-haiku-4.5
Vendor
anthropic
Released
October 2025
Tokenizer
Claude
Input Modalities
text, image, file
Output Modalities
text
Max Output
64,000 tokens
Tool Calling
✓ supported
Structured Output
✓ supported
Reasoning Mode
✓ supported
Vision
✓ accepts images
Audio
no
Moderated
yes

What it costs in practice

Computed from the current $1.00/M input and $5.00/M output rates. Run your own numbers →

JobTokensCost
Summarize a 50-page report 30k in / 1.5k out $0.04
Classify 1,000 customer emails 500k in / 50k out $0.75
A month of a busy support chatbot 5M in / 2M out $15.00

Similar models

Quick answers

How much does Anthropic: Claude Haiku 4.5 cost?
$1.00 per million input tokens and $5.00 per million output tokens.
What is Anthropic: Claude Haiku 4.5's context window?
200,000 tokens, roughly 300 pages of text in a single request.
Does Anthropic: Claude Haiku 4.5 support tool calling?
It supports tool calling, structured output, a reasoning mode.
Can Anthropic: Claude Haiku 4.5 process images?
Yes, it accepts image input.