All Categories
Features Walkthrough
Introduction to Elephas AI Engine

Introduction to Elephas AI Engine

Elephas supports wide range of AI providers from OpenAI to Gemini.

Last updated on October 31, 2025

AI Models & Provider Configuration in Elephas

Elephas gives you the flexibility to use powerful AI models — both built-in and from external providers — depending on your workflow, privacy needs, and subscription plan.

💡

We recommend using Claude Haiku for most writing and brainstorming — it’s highly efficient and credit-friendly.

You can also switch to other models as needed.

Elephas Built-in AI Engine

Elephas, when subscribed through our official website, comes with an inbuilt AI engine backed by OpenAI APIs.

  • Your free account includes a limited number of monthly tokens.
  • Upgrading to a higher plan increases your token limits.
  • Tokens reset monthly and can be tracked under Elephas preferences.
 

Available AI Models (Cloud-Based)

Depending on your subscription plan, you’ll have access to the following AI models:

Provider
Model Name
Description
OpenAI
gpt-4.1 / gpt-4.1-mini / gpt-4o-mini / gpt-5 / gpt-5-mini / gpt-3.5-turbo
Advanced reasoning, creative writing, and analysis
Anthropic
claude-3-haiku / claude-haiku-4.5 / claude-sonnet-4.5
Fast, reliable models for structured and contextual responses
Google
gemma-2-9b-it / gemini-2.0-flash-001 / gemini-2.5-flash / gemini-2.5-flash-lite / gemini-2.5-pro
Ideal for summarization, insight generation, and multimodal input
Alibaba
qwen-2.5-72b-instruct
Great for multilingual and code-based responses
DeepSeek
deepseek-r1-distill-qwen-32b
Optimized for fast inference with lower token costs

Using Elephas Inbuilt Offline Models

Elephas also includes offline AI models, available by default on Apple Silicon (M1 and later) Macs.

These models let you perform chat and Super Brain indexing completely offline — for 100% private and secure operations.

⚡ Chat Models

  • Llama 3.2 3B Instruct
  • Qwen 2.5 1.5B
  • Llama 3.2 1B
  • SmolLM 135M
  • Qwen 3 4B
  • Mistral 7B Instruct v0.3

📘 Indexing Models

  • Multilingual E5 Large
  • Multilingual E5 Small
💡

Offline models ensure your chats and Brain data stay entirely on your Mac — with no internet connection required.

 

🧩 External AI Providers (Bring Your Own Keys)

Elephas also supports integrating your own AI keys from popular providers — giving you control, flexibility, and extended model access.

Supported Providers

 
Notion image

Model Configuration Example

Here is how the configuration page looks,

You can assign different providers for specific tasks:

  • Super Chat → Gemini 2.0 Flash
  • Rewrite → GPT-4o (OpenAI)
  • Continue Writing → Elephas Native Engine

This allows you to mix models to optimize speed, quality, and token usage for your unique workflow.

Notion image

🔗 Related Links

 
Did this answer your question?
😞
😐
🤩
Elephas permission setup Elephas Inbuilt Offline Models for 100% Privacy