Claude Haiku 5.5 vs Claude Sonnet 5.5

Where it leads

• Lower input and output prices, $0.10 and $0.50 vs $2 and $10 per 1M tokens
• Cheaper cached input, $0.01 vs $0.10 per 1M tokens
• Cheaper batch processing, $1 vs $20 for 10M input and 2M output tokens

Best for

• High-volume classification, extraction and routing under the 100,000-token line
• Cost-sensitive fan-out steps where a frontier model is overkill

Watch out for

• Priced by prompt length, and it is a cliff rather than a taper: a request whose prompt passes 100,000 tokens is billed at $0.50 input and $2.50 output for the whole request, five times the short rate
• Cache reads and cache writes both count toward that 100,000-token measurement, so a large cached prefix can tip a request into the higher band on its own
• It is the only current Claude model whose 1M window is not priced uniformly; every other 4.6-and-later model bills a 900k request at the same rate as a 9k one

Where it leads

• Higher Arena Text score, 1477 Elo vs 1425 Elo
• Higher Arena WebDev score, 1770 Elo vs 1589 Elo
• Higher FrontierMath Tiers 1-3 (v2) score, 88.8% vs 75.1%
• Higher GPQA Diamond score, 95.6% vs 89.6%
• Higher SimpleQA Verified score, 46.5% vs 23.8%

Best for

• Production workloads that need frontier-adjacent quality at a mid-tier rate
• Latency-sensitive agent steps inside a larger pipeline

Watch out for

• Same list price as Claude Sonnet 5, so the gain is capability and cache economics rather than a cheaper rate card
• Anthropic positions Opus 5.5 rather than this as the default, so reach for Sonnet 5.5 when latency matters more than peak capability