Claude Opus 5.5 vs Claude Sonnet 5.5

Pyyan / Compare / Claude Opus 5.5 vs Claude Sonnet 5.5

2 of 5

Language Models · verified 29 Sept 2026

Claude Opus 5.5Anthropiccurrent

Claude Sonnet 5.5Anthropiccurrent

Add another3 slots left

Specification	Claude Opus 5.5	Claude Sonnet 5.5
Summary	Fable 5.1's level of work on most tasks, at 40% less to run than Opus 5.	Above Opus 5.5 on Terminal-Bench, at half the input price.
Context	1M	1M
Input ($/Mtok)	$4	$2
Output ($/Mtok)	$20	$10
Licence	Proprietary	Proprietary
Max output	128K	128K
Parameters	Not published	Not published
Released	22 Sep 2026	28 Sep 2026
Model ID	claude-opus-5-5	claude-sonnet-5-5
Category	Language Models	Language Models
Official	Anthropic ↗	Anthropic ↗

Highlighted rows are where these differ.

66.4% on Terminal-Bench 4.0 against 52.3% for Opus 5, on Anthropic's own harness
Cache reads cut 60% to $0.20, and output more than 30% faster than Opus 5
Attempts to work around its boundaries 85% less often than Opus 5, Anthropic reports

Best for agentic coding at flagship quality.

70.6% on Terminal-Bench 4.0, against 66.4% for Opus 5.5 six days earlier
1844 on GDPval-AA, level with Opus 5.5's 1846
More than 30% faster than Sonnet 5, and Anthropic says up to 30% less per task on tokens

Best for agentic coding at volume.