Sage2

Sage2

Turn live talk into structured data and instant actions

A decision and extraction API that turns live conversations into structured data and streaming decisions with calibrated confidence.

Sage2 model page

Sage2 model page

1 / 2

Overview

What is Sage2?

Turn live talk into structured data and instant actions

Sage2 by Levanto Labs is a decision and extraction API for live conversations. It gives developers APIs to extract structured data from natural language and make streaming decisions as conversations evolve: define the decision in plain language, and Sage2 returns a structured answer with a calibrated probability, updates it over WebSocket as more text arrives, and extracts typed values like dates, amounts and durations with their source positions.

  • Decide: Yes/No, Choice, Scale, Tags and Sort decisions with calibrated confidence
  • Stream: WebSocket answers that update as text arrives, in tens of milliseconds
  • Extract: typed values with source positions and confidence
  • Reasoning, web grounding, vision (beta), review thresholds and batch requests
  • Free Playground and docs; data is not used for training

Sage2 has a Free plan at $0 a month including $1 of usage. Paid plans are Developer $14, Starter $49, Pro $99 and Growth $249 a month, each including the same amount in usage credit. Input tokens cost $0.038-$0.05 per million depending on plan ($0.05 on Free/Developer), output $1 per million, and web search $0.10 per search.

Platforms and languages

Sage2 Availability

Platforms

WebApi

Languages

English

Capabilities

Sage2 Key Features

Decide

Define Yes/No, Choice, Scale, Tags or Sort decisions in plain language and get structured answers with calibrated confidence scores.

Stream

WebSocket streaming updates answers as text arrives, with a single decision in tens of milliseconds, about 3x faster than repeated calls.

Extract

Turn natural language into typed values like money, duration and text, with source positions and confidence for every value.

Reasoning, grounding and vision

Toggle reasoning, web search with sources, image Q&A, human-review thresholds and batch requests as needed.

Playground and docs

Test with real conversations in a free Playground or hand the API contract to a coding assistant; existing Sage keys work.

Best for

Who uses Sage2?

Developers building live conversation apps

Extract structured data and make runtime decisions as conversations evolve

Support, sales and learning product teams

Route, escalate and act on changing user intent in real time

Plans and access

Sage2 Pricing

Freemium

Free $0/mo incl. $1 usage; paid plans $14-$249/mo.

View full pricing

Common questions

Sage2 FAQs

Do I need a credit card to start?

No. The Playground is free and the API includes $1 of usage every month without a card.

How is Sage2 different from Sage1?

Sage2 keeps Sage1's decision and calibration abilities and adds a streaming API and structured extraction, at lower prices and higher speed.

How do I know if it fits my use case?

Test it with your real conversations and data in the Playground or API, and check its accuracy and calibration on your own questions.

How is my data handled?

Data is used only to provide the service and is not used for training; see the Privacy Policy for details.

Do I pay for every question?

No. You pay only for input and output tokens; batch and streaming processing carry no extra charge.

Reviews

See what the community thinks and share your experience.

—

Based on 0 ratings

Rating distribution

0
0
0
0
0

Leave a review

Sign in to rate

Community reviews

No reviews

No written reviews yet.

Explore by category

AI Apis Sdks

View all AI Apis Sdks websites
Living Brain preview
Living Brain

A persistent memory layer that lets AI apps and agents remember facts, preferences, and experience across every conversation.

AI Apis Sdks
Sarvam Prism preview
Sarvam Prism

Sarvam AI's open-weight System One decision model that scores fixed options in a single pass and returns calibrated probabilities for enterprise workflows.

AI Apis Sdks
OpenMail preview
OpenMail

An email API that gives every AI agent its own inbox on your domain, with webhooks, attachments as text and MCP support.

AI Apis Sdks
Render Sandboxes preview
Render Sandboxes

Isolated, on-demand Linux sandboxes for running untrusted code and AI agent workflows, billed by actual CPU use.

AI Apis Sdks
UberLLM preview
UberLLM

One OpenAI- and Anthropic-compatible API key for text, image, video and speech AI models, usually below list price

AI Apis Sdks

Explore by category

AI Information Extraction

View all AI Information Extraction websites