Claude Haiku 4.5

Fast

claude-haiku-4-5 · Anthropic · text

Back to Models Back to Docs

Platform

Anthropic

Visit Official Site

API Pricing

Input / 1M

$1/M$0.7/M30% OFF

Output / 1M

$5/M$3.5/M30% OFF

Cache Read / 1M

$0.1/M$0.07/M30% OFF

Overview

Haiku 4.5 is Anthropic's ultra-fast model, offering 200K context, low response latency, and pricing at just 1/3 of Sonnet. Ideal for high-concurrency, low-latency scenarios: customer support, classification, summarization, and lightweight code completion.

Capabilities

200K context; visual understanding; tool calling; prompt caching; JSON mode; Function Calling; fast responses (first token <1s).

Use Cases

High-concurrency customer support; classification and tagging; summarization; lightweight code completion; image recognition; form extraction.

Limitations

Does not generate images; not recommended for complex reasoning tasks; shorter context (200K); no Extended Thinking; max_tokens capped at 8K.

Examples

Claude Code

Anthropic official CLI coding agent. Multi-file edits, tool use, long-context reasoning.

Terminal

export ANTHROPIC_AUTH_TOKEN="YOUR_API_KEY"
export ANTHROPIC_BASE_URL="https://api.robovai.com"
export ANTHROPIC_MODEL="claude-haiku-4-5"
claude

Context Window

200K ctx

Thinking Mode

Not supported

Overview

Claude Haiku 4.5, 200K context, extreme speed and cost-efficiency, ideal for high-concurrency lightweight tasks.