Inflection AI

Inflection AI is a foundation-model company behind the Pi personal AI assistant and the Inflection 3.0 family of empathetic large language models (Pi, Productivity). The Inflection Developer API serves these proprietary models for enterprise integration with cloud and on-premise deployment options.

6 APIs 0 Features

AILLMPersonal AIPiFoundation ModelsEmpathetic AI

APIs

Inflection Chat API

Generates chat-style completions against Inflection's empathetic foundation models (inflection_3_pi, inflection_3_productivity, Pi-3.1) with streaming responses.

Inflection Models API

Lists the available Inflection 3.0 models (inflection_3_pi, inflection_3_productivity, Pi-3.1) with metadata.

Inflection Playground

Browser-based playground for prototyping prompts and evaluating Inflection 3.0 models.

Inflection Usage API

Surfaces usage metrics and rate-limit consumption for the developer account.

Pi Personal AI

Pi consumer chat product (web, iOS, Android, WhatsApp, SMS, Telegram) - empathetic personal AI assistant. Not a developer surface; access is via consumer apps.

Inflection for Enterprise

Enterprise deployment of Inflection 3.0 models including a purpose-built on-premise server option for data locality and TCO optimization. Contact sales.

Collections

Inflection Inference API

OPEN

Pricing Plans

Inflection Plans Pricing

3 plans

PLANS

Rate Limits

Inflection Rate Limits

3 limits

RATE LIMITS

FinOps

Inflection Finops

FINOPS

Resources

👥

GitHubOrganization

Sources

opencollection: 1.0.0
info:
  name: Inflection Inference API
  version: 1.0.0
items:
- info:
    name: 'Status '
    type: http
  http:
    method: GET
    url: '{{baseUrl}}/status'
  docs: 'Status '
- info:
    name: Chat Completions
    type: http
  http:
    method: POST
    url: '{{baseUrl}}/v1/chat/completions'
    body:
      type: json
      data: '{}'
    auth:
      type: oauth2
      flow: resource_owner_password_credentials
      accessTokenUrl: token
      credentials:
        clientId: '{{clientId}}'
        clientSecret: '{{clientSecret}}'
  docs: 'Chat completions endpoint for Inflection hosted models.


    ⚠️ This endpoint is under active development to enable all Inflection model and config types.'
- info:
    name: Chat Attributes
    type: http
  http:
    method: POST
    url: '{{baseUrl}}/v1/chat/attributes'
    body:
      type: json
      data: '{}'
    auth:
      type: oauth2
      flow: resource_owner_password_credentials
      accessTokenUrl: token
      credentials:
        clientId: '{{clientId}}'
        clientSecret: '{{clientSecret}}'
  docs: 'Get attribute scores for the most recent assistant message in a chat completions style request.


    ⚠️ Experimental endpoint for evaluating message attributes.'
- info:
    name: Embeddings
    type: http
  http:
    method: POST
    url: '{{baseUrl}}/v1/embeddings'
    body:
      type: json
      data: '{}'
    auth:
      type: oauth2
      flow: resource_owner_password_credentials
      accessTokenUrl: token
      credentials:
        clientId: '{{clientId}}'
        clientSecret: '{{clientSecret}}'
  docs: Embeddings endpoint for Inflection hosted models.
- info:
    name: Configs
    type: http
  http:
    method: GET
    url: '{{baseUrl}}/v1/discovery/configs'
    auth:
      type: oauth2
      flow: resource_owner_password_credentials
      accessTokenUrl: token
      credentials:
        clientId: '{{clientId}}'
        clientSecret: '{{clientSecret}}'
  docs: 'Get all available configs (for use in the chat completion endpoint `model` field) for an organization.


    Returns each config''s `name` (aka `model`), `alias` (for UI display), `default_parameters`, and `description`.'
- info:
    name: Legacy Chat Completions
    type: http
  http:
    method: POST
    url: '{{baseUrl}}/external/api/inference/openai/v1/chat/completions'
    body:
      type: json
      data: '{}'
    auth:
      type: oauth2
      flow: resource_owner_password_credentials
      accessTokenUrl: token
      credentials:
        clientId: '{{clientId}}'
        clientSecret: '{{clientSecret}}'
  docs: 'Chat completions compatible endpoint for legacy Inflection models.


    Provides a compatibility chat completions endpoint to talk with any Inflection model.'
- info:
    name: Legacy API Inference Chat
    type: http
  http:
    method: POST
    url: '{{baseUrl}}/external/api/inference'
    params:
    - name: origin_endpoint
      value: ''
      type: query
    body:
      type: json
      data: '{}'
    auth:
      type: oauth2
      flow: resource_owner_password_credentials
      accessTokenUrl: token
      credentials:
        clientId: '{{clientId}}'
        clientSecret: '{{clientSecret}}'
  docs: Calls an Inflection model using the legacy chat API.
- info:
    name: Legacy API Streaming
    type: http
  http:
    method: POST
    url: '{{baseUrl}}/external/api/inference/streaming'
    params:
    - name: origin_endpoint
      value: ''
      type: query
    body:
      type: json
      data: '{}'
    auth:
      type: oauth2
      flow: resource_owner_password_credentials
      accessTokenUrl: token
      credentials:
        clientId: '{{clientId}}'
        clientSecret: '{{clientSecret}}'
  docs: Calls an Inflection model using the legacy chat streaming API.
bundled: true