PrismML: Ternary Bonsai 2 9B vs NVIDIA: Nemotron 3.5 Super VL

For a sample job of 1 million input tokens and 250,000 output tokens, Ternary Bonsai 2 9B costs $0.11 and Nemotron 3.5 Super VL costs $0.28, so Ternary Bonsai 2 9B is about 2.4 times cheaper for the same work.

Nemotron 3.5 Super VL reads up to 1,048,576 tokens at once, roughly 1,573 pages of text, against about 393 pages for Ternary Bonsai 2 9B.

Only Nemotron 3.5 Super VL offers video input.

Across the 18 tasks where they differ, Ternary Bonsai 2 9B scores higher on 4 (Regex Writing, JSON Extraction, Bulk Data Labeling and others), Nemotron 3.5 Super VL on 14 (SQL Generation, Code Review, Code Completion and others).

Pick Ternary Bonsai 2 9B if you need lower cost; pick Nemotron 3.5 Super VL if you need long documents, video input or the stronger all-round task scores.

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