Inception's previous-generation diffusion llm from September 2026, text-only, with tunable reasoning and a 260K context window. it retires on 2026-11-04; Mercury 3, the current generation, has lower input and output rates.
pick it if
you need fast text generation with reasoning and schema-aligned JSON
your prompts repeat, so cached input bills at a tenth of input
skip it if
Mercury 3 costs less on both rates and has a 1M window
Google's Flash-Lite tier Gemini model for high-volume agentic work, subagent tasks, document parsing and simple data extraction. it is the cheapest Gemini in the catalog, and Google positions it as a replacement for Gemini 2.5 Flash.
pick it if
you want the cheapest Gemini in the catalog for high-volume calls
you send images, video, audio or PDFs alongside text
you resend long prompts; cache hits bill at 10% of input
skip it if
your coding work goes beyond the simple tasks Google targets
you need image or audio output; it returns text only
faq
which is cheaper, Mercury 2.5 or Gemini 3.5 Flash Lite?
Mercury 2.5 lists lower input and output prices. on a coding-agent session (2M tokens in, 60,000 out) Mercury 2.5 costs $0.45 and Gemini 3.5 Flash Lite $0.75.
which has the larger context window?
Gemini 3.5 Flash Lite offers the larger context window. Mercury 2.5: 260k tokens; Gemini 3.5 Flash Lite: 1M tokens.
which one accepts images?
Gemini 3.5 Flash Lite supports vision; Mercury 2.5 does not.
try either model
switch between models inside the same octomind session.