Anthropic: Claude Opus 5.5 (batch) vs PrismML: Ternary Bonsai 2 27B

For a sample job of 1 million input tokens and 250,000 output tokens, Ternary Bonsai 2 27B costs $0.20 and Claude Opus 5.5 (batch) costs $4.50, so Ternary Bonsai 2 27B is about 22.5 times cheaper for the same work.

Claude Opus 5.5 (batch) reads up to 1,000,000 tokens at once, roughly 1,500 pages of text, against about 393 pages for Ternary Bonsai 2 27B.

Across the 18 tasks where they differ, Claude Opus 5.5 (batch) scores higher on 13 (SQL Generation, Code Review, Code Refactoring and others), Ternary Bonsai 2 27B on 5 (Code Completion, Regex Writing, JSON Extraction and others).

Pick Claude Opus 5.5 (batch) if you need long documents or the stronger all-round task scores; pick Ternary Bonsai 2 27B if you need lower cost.

Costs use current list prices per million tokens; page counts assume about 0.75 words per token and 500 words per page.