Claude Haiku 4.5
Claude Haiku 4.5 is a closed AI model from Anthropic with an index score of 137 (GPT-4 = 100), rank 108 of 217.
137121–153Index score#108 of 217 models
$1.0/ $5.0Price per million tokensinput / output · via OpenRouter
200KContext windowtokens it reads at once
15 Oct 2025ReleasedReplaced by Claude Haiku 5.5
Where it sitson the frontier
Gemini 2.5 Pro reached this score 7 months earlier.
Test resultsby area · tick = best by any model
Reasoning & knowledgeevidence
Mathsevidence
Codingevidence
Agentsevidence
Professional workevidence
Novel problemsevidence
Long tasksno evidence yet
Third-party results only, at the model’s best standard setting. Hover a bar for who ran it.
What others sayrank on 10 leaderboards
The Model Indexthird-party tests, our method#108of 217 · 137
LMArena · Codingcoding questions in chat#95of 357 · 1482LMArena · Instruction followingfollowing detailed instructions#105of 362 · 1415LMArena · Long promptslong, complex requests#99of 340 · 1436LMArena · Creative writingcreative writing#111of 360 · 1387LMArena · Textoverall chat quality, judged by people#119of 362 · 1414LMArena · Mathsmaths questions in chat#132of 346 · 1402Epoch Capabilities Indexoverall capability from many benchmarks#107of 231 · 142.8LMArena · WebDevbuilding web apps and front ends#111of 133 · 1330LMArena · Documentsworking with long documents#34of 40 · 1420Best attop three among current models
Replaced by Claude Haiku 5.5; it no longer leads any category among current models.
Where to use itprices checked 18 Nov 2026
API providers 4 · per 1M tokens, in / out
Host prices from OpenRouter; going direct can differ.
Facts and sources
Released15 Oct 2025
Announcementanthropic.com ↗
Release entryClaude Haiku 4.5
Scored asClaude Haiku 4.5
Output speed–
Hugging Faceclosed weights