Inception: Mercury 2

Mercury 2 is a text-in, text-out model from Inception with a 128,000-token context window and a maximum of 50,000 completion tokens per response. It supports tool use and reasoning, which makes it suitable for multi-step workflows and agentic tasks. Structured output support is not confirmed, so developers who depend on guaranteed JSON schemas should verify compatibility before committing. At $0.25 per million input tokens and $0.75 per million output tokens, Mercury 2 sits in the budget-to-mid tier on price. Its blended benchmark score of 41.2 comes from a single benchmark, so that figure is thin evidence for broad capability comparisons. Teams running high-volume, tool-augmented pipelines on a tight budget may find the pricing worth a trial, but buyers who need confidence across diverse task types should wait for broader benchmark coverage before treating that score as representative.