Google Sets Gemini Decision 1 Input Price at $0.10 per Million Tokens
In Google's self-test, Gemini Decision 1 led TypeSafe's Jev 1.14 on five of seven shared benchmarks. Output tokens carry no fee.

Google set Gemini Decision 1's input price at $0.10 per million tokens, with no fee for output, a day after it introduced the model through Google AI Studio on Nov. 11.
TypeSafe's Jev 1.14 has an input price of $0.042 per million tokens. Gemini Decision 1 therefore has the higher listed input rate, while its output tokens are not billed.
In Google's self-test, Gemini Decision 1 scored 92.2% on JevBench, compared with 90.0% for Jev 1.14. It led on five of the seven benchmarks shared by the models.
Gemini Decision 1 handles questions with fixed answer choices by scoring options in a single pass instead of generating text token by token. The API accepts text, images, audio, video and PDFs, and can process as many as 1,000 questions in one request. Multiple-choice questions can include up to 255 options, and each answer includes a confidence score. The model has a 256,000-token context window and end-to-end latency of about 150 to 250 milliseconds.