modelgrep
A

Anthropic: Claude Haiku 4.5

anthropic/claude-haiku-4.5

76th smartest of 178ReasoningToolsJSONVision
Use via OpenRouter ↗
Intelligence
31.0
76th of 178
Design Elo
1182
asciiart
Speed
82
90th fastest
Latency
521ms
first token
Input price
$1.00
206th cheapest
Context
200K
64K max out

How it compares

Smarter than57%
of all ranked models
Faster than69%
of all ranked models
Cheaper than31%
of all ranked models

Overview

Claude Haiku 4.5 is Anthropic’s fastest and most efficient model, delivering near-frontier intelligence at a fraction of the cost and latency of larger Claude models. Matching Claude Sonnet 4’s performance...

Benchmarks

independent · via OpenRouter
Artificial Analysis56th percentile
Intelligence Index
31.0
Coding Index
29.6
Agentic Index
32.6
GPQA Diamond
65%
Humanity's Last Exam
4%
SciCode
34%
Tau²-Bench (agentic)
33%
Design Arena · Elo18,479 tournaments
asciiart
1182
Data Viz
1169
Website
1166
codecategories
1166
Game Dev
1165
UI Component
1156
3D
1155
svg
1082

Providers & pricing (5)

ProviderIn $/MOut $/MUptime
Amazon Bedrock$1.00$5.00
Google$1.00$5.00
Google$1.00$5.00100%
Anthropic$1.00$5.00100%
Amazon Bedrock$1.00$5.0099.9%

Specifications

Context window200K
Max output64K
Knowledge cutoff
Input modalitiestext, image, file
Output modalitiestext
Prompt caching
Cache read price$0.100/M
ModeratedYes

Claude Haiku 4.5 FAQ

How much does Claude Haiku 4.5 cost?

Claude Haiku 4.5 costs $1.00 per million input tokens and $5.00 per million output tokens via OpenRouter, making it 206th cheapest of 298 paid models.

How smart is Claude Haiku 4.5?

Claude Haiku 4.5 scores 31.0 on the Artificial Analysis Intelligence Index, ranking 76th of 178 benchmarked models, with a GPQA Diamond score of 65%.

How fast is Claude Haiku 4.5?

Claude Haiku 4.5 generates around 82 tokens per second with 521ms time-to-first-token (p50), the 90th fastest tracked model.

What is Claude Haiku 4.5's context window?

Claude Haiku 4.5 supports a 200K-token context window and can output up to 64K tokens. It accepts text, image, file input.

Compare head-to-head