Anthropic
Claude Haiku 4.5
Release Date 2025-10-15(Knowledge cutoff 2025-02)200K IN · 64K OUTProprietary ModelAPI availableClaude Free and up
Claude Haiku 4.5 is Anthropic's fastest and most cost-efficient model, delivering near-frontier intelligence at a fraction of the cost of larger Claude models. It matches Claude Sonnet 4's coding performance at one-third the cost and over twice the speed, scoring 73.3% on SWE-bench Verified — placing it among the world's top coding models. With support for extended thinking, tool use, computer use, and a 200K-token context window, it is ideal for real-time applications, parallelized sub-agents, and high-volume deployments.
API cost
Official price
$1IN$5OUT
SourceAnthropic official docs
Real-world usage
Outside weekly top 50
243B8/24
240B8/31
SourceOpenRouter
Output Speed
Fastper 2,000 tokens
68tok/s
First Output 0.9stotal: 30.3s
SourceOpenRouterLast updated
User Comparisons
Response Speed
First response 1.7sTotal 6.0s
88 tok/s
SourceAIB Inc.Last updated
Expert Performance Evaluation
Overall performance
All overall performance AI rankings16.9Reasoning effort: Thinking141480.0%
By effortNone 15.4Thinking 16.9
Automation (agents)
All automation AI rankings9.3%Reasoning effort: Thinking$45954.7%
Documents and office work
All documents and office work AI rankings74.3%Reasoning effort: Thinking142031.0%60.6%28.3%10.6%0.8%32.7%142046.9%43.0%
By effortNone 49.7%Thinking 74.3%
Research and fact-checking
All research and fact-checking AI rankings141813.2%45.5%12.8%9.8%90.2%59.0
Coding
All coding AI rankings0.0%Reasoning effort: Thinking148165344.2%27.3%43.9132961.5%42.2%45.4%146435.5%
Math and science
All math and science AI rankings10.4%Reasoning effort: Thinking0.0%140067.2%66.7%83.783.7%1430143396.4%5.9%2.1%47.7%4.0%8.0%73.6%30.9
By effortNone 4.2%Thinking 10.4%
Image understanding
All image understanding and analysis AI rankingsCan it draw a floor plan from a few photos of a home?Blueprint-Bench 2Bottom 11%#27
0.0%
Specialized fields
1366136183.0%144114511425141613.7%31.2%87.0%66.9%73.3%60.4%83.0%79.7%93.5%79.6%77.0%64.0%33.3%26.0%9.0%77.0%8.8%82.0%67.0%62.0%75.0%76.0%36.3%78.8%88.7%73.7%
See each evaluation’s detail page for its source and last updateData sources & removal requests
Related Info
PrivateAI Native
Anthropic🇺🇸 San Francisco2021ARR $30B+$380B~1,200