Claude Haiku 5.5 vs Claude Opus 5.5

Where it leads

• Lower input and output prices, $0.10 and $0.50 vs $4 and $20 per 1M tokens
• Cheaper cached input, $0.01 vs $0.20 per 1M tokens
• Cheaper batch processing, $1 vs $40 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 GPQA Diamond score, 90.6% vs 89.6%
• Higher FrontierMath Tiers 1-3 (v2) score, 91.2% vs 75.1%
• Higher SimpleQA Verified score, 72.2% vs 23.8%

Best for

• Long-running agentic coding and knowledge work as the general-purpose default
• Cache-heavy agent loops where the 0.05x read multiplier compounds

Watch out for

• The API effort default is medium here, not high as on Claude Fable 5.1 and Claude Sonnet 5.5, so a like-for-like comparison has to set effort explicitly
• The 4.7-generation tokenizer produces roughly 30% more tokens for the same text than Claude Sonnet 4.6 and earlier, so per-token rates do not compare directly across generations