Productivity & writingDocumented overviewLast checked Oct 2, 2026

iGPT: Pricing, Features & Best Use Cases

Context engine that indexes email and work data, offered as a chat app and a developer API.

Who iGPT fits

iGPT is a good fit for developers who need cited, schema-shaped answers from email and attachment data through one API, and for individuals or teams who want a context-aware chat with a free starting tier. Based on the vendor's documentation, not on a hands-on test.

Best for: Developers building email-aware agents

Also a good fit for

  • Teams needing org-wide AI search

Consider another tool if free tier has lower usage limits

Or if

  • Team plan has a 3-seat minimum
  • Agent builder and extra datasources are not in Free

iGPT pricing and usage limits

Prices checked Oct 2, 2026 on the official US pricing page. Official pricing ↗ Prices and limits change often.

Free access and trial conditions

iGPT Chat has a Free plan at $0 with usage limits.

$10 in free credits to explore the platform (API).

Paid plans: what changes as you upgrade

Free chat

Free source (opens in a new tab)

Month

For: Everyday context-aware chat

Limits

  • Limits apply based on usage
  • Email connection only
1 more
  • Lower usage limits

Includes

  • CEF-Normal context reasoning
  • Real-time web search
1 more
  • Web, desktop and mobile access (per comparison table)

No API

Pro chat

$25/mo source (opens in a new tab)

Month

$25 if billed monthly; $20 per month with annual subscription discount.

For: Deep context and critical work

Limits

  • Higher usage limits
  • Full datasource stack

Includes

  • Powered by CEF-High
  • Faster responses during peak usage
1 more
  • Priority access to new capabilities

No API

Team chat

$35 source (opens in a new tab)

Seat/month

$35 per seat per month billed monthly; $30 per seat per month billed annually.

For: Teams and organizations

Limits

  • Minimum 3 seats
  • Custom, org-defined usage limits

Includes

  • Role-aware assistants and agents
  • Admin console and RBAC
2 more
  • Agent builder
  • SOC 2 Type II

No API

Context Engine API (CEF-4 Normal) api

$1.1 per 1M input tokens source (opens in a new tab)

Usage

For: Developers building email-aware agents

Limits

  • Output $6.50 per 1M tokens
  • $10 free credits to explore
2 more
  • CEF-4 High: $3.50 input, $21 output per 1M
  • CEF-5 High: $7 input, $42 output per 1M

Includes

  • Single unified endpoint
  • Structured output schemas
1 more
  • Citations to source messages

API included

Enterprise enterprise

Custom pricing source (opens in a new tab)

Custom

For: Organizations needing governance

Limits

  • Custom organizational deployment

Includes

  • Cloud, hybrid or private infrastructure
  • RBAC and audit controls
1 more
  • No-code agents on company knowledge

Usage caps, credits and extra costs

API usage is billed per million input and output tokens by tier.

What is iGPT, and how does it work?

iGPT is a context engine for work. It indexes emails and attachments and returns structured, cited outputs through a single API, and also offers iGPT Chat for context-aware assistance. Plans include a free chat tier, Pro and Team, plus pay-as-you-go API pricing and custom enterprise deployments.

iGPT homepage — screenshot of the productivity & writing AI tool, Oct 2, 2026
iGPT's official site: how the product presents itself. igpt.ai, captured Oct 2, 2026

What you provide and what you get back

How iGPT works: what you provide, where it runs and what you get backYOU PROVIDEYour inputConnected email accounts and atta…Calendar, documents and tools via…Natural-language questionsOptional JSON output schemaPRODUCTIVITY & WRITINGiGPTWeb, Desktop, Mobilefrom $25/mo (Pro)free tier availableaccount requiredchecked Oct 2, 2026YOU GET BACKThe resultCited answersStructured JSON matching a schemaDrafts and summariesAgent actionsinput → iGPT → output · from the vendor's own pages, no hands-on test unless stated
What you give iGPT and what comes back, from its documentation

You provide

  • Connected email accounts and attachments
  • Calendar, documents and tools via OAuth
  • Natural-language questions
  • Optional JSON output schema

You get back

  • Cited answers
  • Structured JSON matching a schema
  • Drafts and summaries
  • Agent actions

Where you can use it

Platforms
Web, Desktop, Mobile, API
Account
Required
Datasources
Email, calendar, documents, tools via OAuth
API model tiers
CEF-4 Normal, CEF-4 High, CEF-5 High
Platforms
Web, desktop app, iOS, Android (comparison table)
Integrations
API, SDK, MCP, LangChain
Latency claim
~200ms retrieval, ~3s to first token (vendor-reported)

Best use cases for iGPT

Scenarios are worked out by our editors from the documentation; the actual result depends on your input and plan.

Extract action items from email threads

Who
Developers and operations teams
You provide
A question like open action items from yesterday's board meeting plus a JSON schema
Expected result
Structured JSON of action items with owners and due dates
Plan needed
Context Engine API
Limitation
Needs API integration and a connected mailbox source (opens in a new tab)

Give agents email context via API

Who
AI agent builders
You provide
Connected user email via the API
Expected result
Agents that draft, prioritize and act on email with thread context
Plan needed
Context Engine API
Limitation
Token costs vary by CEF tier source (opens in a new tab)

Chat over connected email data

Who
Professionals and teams
You provide
Connected email and documents in iGPT Chat
Expected result
Context-aware answers and drafts drawn from real threads
Plan needed
Free, Pro or Team
Limitation
Additional datasources are not in Free source (opens in a new tab)

iGPT features and their limits

iGPT homepage — screenshot of the productivity & writing AI tool, Oct 2, 2026
Feature overview on the official site. igpt.ai, captured Oct 2, 2026

Single unified API endpoint

Runs retrieval, context shaping and reasoning in one natural-language request across connected datasources. Removes the need for separate vector stores and prompt chains.

Access: APILimit: Billed per token by CEF tier. source (opens in a new tab)

Structured JSON output schemas

Constrains answers to a user-supplied JSON schema so responses are ready for automation. Feeds results straight into code.

Access: APILimit: You must design and maintain the schema. source (opens in a new tab)

Email and attachment indexing

Indexes email messages and extracts text, data and structure from documents, PDFs and spreadsheets, rebuilding threads across participants and time. Answers reflect how conversations unfolded.

Access: Email connection on all chat plansLimit: Additional datasources are not included in Free. source (opens in a new tab)

Python SDK and playground

Provides an SDK and live playground so requests and responses can be tested quickly, with MCP and LangChain integration listed. Speeds first integration.

Access: APILimit: Docs page was not available when fetched. source (opens in a new tab)

iGPT pros and cons

Where iGPT is strong

Limitations to check before paying

How to use iGPT for extracting action items from email threads

Set up iGPT

  1. Open igpt.ai and choose Get Free or Get your API key.
  2. Connect an email account through OAuth.
  3. For API use, create an API key and set IGPT_API_KEY.
  4. Install the Python SDK or call the REST endpoint.

Example: extracting action items from email threads

  1. Ask for open action items from yesterday's board meeting.
  2. Pass a JSON schema with title, owner and due_date.
  3. Choose a quality tier such as cef-1-normal.
  4. Read the structured response with citations.

Review and export the result

  1. Open the cited source messages and confirm each action item.
  2. Compare owners and dates against the original thread.
Example prompt suggested by our editors (not a recorded output of the product)
List open action items from yesterday's board meeting with owner and due date.

iGPT integrations and API

Native integrations and export formats

Export: JSON

API access and separate billing

iGPT has no public API in its official documentation. Docs URL returned HTTP 404; endpoint example at api.igpt.ai/v1/recall/ask/.

Privacy, data use and commercial rights

Summarized from the vendor's current policies; this is not legal advice and a vendor's badge is not an independent audit.

How your inputs and outputs are handled

Connects through OAuth, searches your emails and files, and mirrors source permissions so users only see what they can already access.

Site states customer data is not used for model training.

Designed for zero data retention at inference; disconnected sources or deleted files are removed from the index.

Commercial use and plan restrictions

Cloud, hybrid or private deployment, audit logs and SOC 2 Type II are listed for Team and Enterprise.

Controls for business users

  • RBAC
  • Audit logs
  • SOC 2 Type II
  • Hybrid and on-prem deployment

Policies: igpt.ai/privacy-policy/

iGPT alternatives by need

"Better suited" is based on documented features and pricing, not on a head-to-head test.

ChatGPT: better suited to general AI assistant

Claude: better suited to general AI assistant

Otter.ai: better suited to meeting notes and transcripts

Notion AI: better suited to aI inside a workspace

Alternatives
AlternativeWhen to choose itPricingLimitation
ChatGPTyou want broad chat without email-specific retrieval— source (opens in a new tab)Does not index your email threads natively
Claudeyou want a general assistant with strong writing— source (opens in a new tab)No dedicated email-context API with citations
Otter.aiyour key context is meetings rather than email— source (opens in a new tab)Not an email and attachment context engine
Notion AIyour knowledge lives in Notion— source (opens in a new tab)Less focus on email thread reconstructionCompare

Explore iGPT further

Related to iGPT

iGPT user reviews

No user reviews yet.

Used iGPT? Be the first to share your experience.

Write a review

Reviews are checked by a moderator before they appear. Your email is never published. Plain text only.

Your rating (required)

Show what you made or what went wrong. Images are published only if the review is approved; location data is removed.

    iGPT FAQ

    What does iGPT Pro cost?

    iGPT Pro is $25 per month billed monthly, or $20 per month with an annual subscription. source (opens in a new tab)

    What are the iGPT Team seat requirements?

    Team is $35 per seat per month billed monthly or $30 annually, with a minimum of 3 seats. source (opens in a new tab)

    How is the iGPT Context Engine API priced?

    CEF-4 Normal costs $1.10 per 1M input tokens and $6.50 per 1M output tokens; CEF-4 High and CEF-5 High cost more, and $10 in free credits is offered. source (opens in a new tab)

    Does iGPT train on my emails?

    No; iGPT states emails, documents and chat history are never used for model training. source (opens in a new tab)

    Which deployment options does iGPT support?

    iGPT supports managed cloud, hybrid with data in your VPC, and fully on-prem deployment. source (opens in a new tab)

    Sources and review methodology

    Official sources and verification dates

    Not verified: Developer docs (HTTP 404); Exact usage limits per plan; Commercial-use terms; US availability not stated.

    How we documented and maintain this page

    Every fact about iGPT above comes from the 3 official pages listed here, last compared on Oct 2, 2026; we have not run iGPT ourselves, so it has no editorial score. How we check and re-check pages.

    Article updated Oct 2, 2026; prices checked Oct 2, 2026. Outbound links to vendors are plain links; we do not earn commissions. How we review · Report an error