llm.ing

Claude Haiku 5.5

AnthropicOpenRouter ↗Artificial Analysis ↗

Fast Claude model for responsive assistance, classification, and lightweight agents

Benchmarks

BenchmarkCategoryScoreVariantProvenanceSourceObserved
AA Intelligence Indexintelligence36.5—mirroropenrouterNov 4, 2026
29.9high-effortindependentaaNov 4, 2026
34.6xhigh-effortindependentaaNov 4, 2026
Design Arena — 3dpreference1349—mirroropenrouterOct 16, 2026
1310—mirroropenrouterNov 19, 2026
Design Arena — asciiartpreference1208—mirroropenrouterNov 18, 2026
Design Arena — codecategoriespreference1272—mirroropenrouterNov 19, 2026
Design Arena — datavizpreference1286—mirroropenrouterOct 16, 2026
1283—mirroropenrouterOct 17, 2026
1275—mirroropenrouterNov 18, 2026
Design Arena — gamedevpreference1318—mirroropenrouterOct 16, 2026
1316—mirroropenrouterOct 17, 2026
1306—mirroropenrouterNov 19, 2026
Design Arena — pdfpreference1215—mirroropenrouterNov 19, 2026
Design Arena — svgpreference1196—mirroropenrouterNov 19, 2026
Design Arena — uicomponentpreference1300—mirroropenrouterOct 16, 2026
1299—mirroropenrouterOct 17, 2026
1265—mirroropenrouterNov 18, 2026
Design Arena — websitepreference1249—mirroropenrouterNov 18, 2026
LMArena WebDevpreference1585high-effortcrowdlmarenaOct 13, 2026
Median output speedspeed168 tok/shigh-effortindependentaaNov 19, 2026
182 tok/sxhigh-effortindependentaaNov 19, 2026
Median time to first tokenspeed26.05shigh-effortindependentaaNov 18, 2026
69.19sxhigh-effortindependentaaNov 19, 2026

Includes index data from Artificial Analysis.

Providers

Live operational stats per hosting endpoint, via OpenRouter.

ProviderQuantContext$/1M in · outUptime 24hLatencyThroughput
Amazon Bedrock—1M$0.1 · $0.5100.00%——
Anthropic—1M$0.1 · $0.5100.00%——
Azure—1M$0.1 · $0.5100.00%——
Claude Platform on AWS—1M$0.1 · $0.5100.00%——
Google—1M$0.1 · $0.5100.00%——