Claude Haiku 5.5 is Anthropic's small, fast model for high-volume, cost-sensitive work such as summarization, subagents, and browser use. It succeeds Claude Haiku 4.5 with stronger coding, computer use, and...
Tested on 20 benchmarks · BenchGecko score 62.3. Top scores: OTIS Mock AIME 2024-2025 (98.9%), Proofbench (89.0%), GPQA diamond (86.2%).
Mock AIME (American Invitational Mathematics Exam) problems from OTIS. Tests mathematical competition performance.
Graduate-level science questions written by PhD experts. Diamond subset contains questions where experts disagree, testing deep understanding.
Tactical chess puzzles testing pattern recognition and multi-move calculation. Measures strategic reasoning ability.
Humanitys Last Exam. Expert-level questions spanning all academic disciplines, designed to be the hardest knowledge test for AI.
- Typemultimodal
- Context1.0M tokens (~500 books)
- ReleasedOct 2026
- LicenseProprietary
- StatusActive
- Cost / Message~$0.001
Frequently Asked Questions
Related Models
Qwen3.6 27B · Alibaba QwenQwen3.6 Plus · Alibaba QwenKimi K2.7 Code · moonshotaiGrok 4.5 · xAIPaLM 2-S · UnknownKey facts · as of 2026-10-15
- Claude Haiku 5.5 by Anthropic. BenchGecko score 62.3, rank 76 of 319 scored models (normalized average of public benchmark scores).
- List price $0.10 input · $0.50 output per 1M tokens (as of 2026-10-15).
- Sold by 10 providers (as of 2026-10-15): Amazon Bedrock $0.10 in / $0.50 out · Anthropic $0.10 in / $0.50 out · Azure $0.10 in / $0.50 out · Claude Platform on AWS $0.10 in / $0.50 out · Google $0.10 in / $0.50 out · and 5 more. Every provider
- Gecko Tests: Who Are You B (Knows who made it).
How to cite · data as of 2026-10-15
Claude Haiku 5.5 · benchmarks, pricing and providers. BenchGecko, data as of 2026-10-15. https://benchgecko.ai/model/claude-haiku-5-5
Credit "Source: BenchGecko" with a link. Prices per provider and Gecko Tests are BenchGecko data (CC BY 4.0); benchmark scores keep their original source, listed in the JSON. JSON · llms.txt · MCP