Claude Haiku 4.5
Haiku 4.5 matches Sonnet 4 performance on coding and agent benchmarks at a fraction of the cost, making it ideal for real-time and high-volume products.
Anthropic · Claude
Claude Haiku 4.5
Fastest, lowest-cost model in the Claude family
Haiku 4.5 matches Sonnet 4 performance on coding and agent benchmarks at a fraction of the cost, making it ideal for real-time and high-volume products.
Key features
Claude Haiku 4.5 · AI Models
Context window
200K tokens
Max output
64K tokens
Released
October 15, 2025
Pricing
$1 input / $5 output per 1M tokens
Key features
- Matches Sonnet 4 quality on coding and agentic tasks.
- 73.3% on SWE-bench Verified.
- 200K context window with 64K max output.
- Lowest pricing in the Claude lineup.
Best for
Pick Haiku 4.5 for latency-sensitive applications, embedded coding agents and any high-throughput chat where cost matters.
Frequently Asked Questions
How fast is Claude Haiku 4.5?
Haiku 4.5 is the fastest model in the Claude family, optimised for real-time applications and low-latency agent loops.
Open Chat
Haiku 4.5 matches Sonnet 4 performance on coding and agent benchmarks at a fraction of the cost, making it ideal for real-time and high-volume products.