Hard, multi-step reasoning
Problems where each step depends on the one before: analysis, planning, trade-offs. Auto sends long or tool-heavy requests here.
ModelsDeep · Kavi
Deep. For the work that has to be right.
Kavi is the Deep size: the largest context window, the largest output budget and the deepest reasoning of the three. For long documents, hard problems and tool-heavy work.
256K
tokens of context, the same on every plan.
8K
tokens of output in one reply, at most.
$0.76
per 1M input tokens through the API.
$1.52
per 1M output tokens through the API.
Text in, text out. Rates are USD per 1M tokens, metered per request. API prices exclude GST. 18% GST is added for India-billed accounts (input credit available). On the web, every plan includes Kavi.
Built for
What Kavi is for.
Long documents in one pass
256K tokens of context, so a long report, a codebase or a contract fits at once. Ask, summarise and cross-reference without chunking.
Writing and code that hold the thread
Long reports, refactors and prompts with many constraints, with the whole context in view and the largest reply budget of the three.
Pick by task
Three sizes. Auto picks one for each message; you can always choose.
PICK IT WHEN
The answer must be right, the document is long, or the reasoning is layered.PICK BALANCED WHEN
The task is everyday work: writing, summarising, answering questions, most coding. It is the default.About VaaniFast· MiraPICK FAST WHEN
Speed matters more than depth: chat, classification, real-time responses.About MiraON EVERY PLAN
Bring the hard problem.
Free on the web, no card required. Per token through the API, at the rates above.