Skip to main content

Language Models

AI-Public unlocks access to the most advanced models for all employees of your organization. We also closely monitor new versions and models entering the market and adapt our application accordingly. This way, you can always use the best and newest AI models.

Thinking Mode​

A powerful feature within AI-Public is thinking mode. This allows you to switch between a fast model for daily tasks and a powerful "thinking model" for complex questions.

How Does It Work?​

When an administrator has configured a thinking model with a fast alternative, you'll see a 💡 icon next to the text field. By clicking it, you toggle between:

ModeIconModelSuitable for
Off (default)💡Fast modelQuick questions, simple tasks
On💡 (active)Thinking modelComplex analysis, deep reasoning

Benefits​

  • 🚀 Faster: The fast model responds quicker to simple questions
  • 💰 Cheaper: Fast models use fewer tokens
  • 🧠 Flexible: Switch to thinking model when you need deeper analysis
  • ✨ Best of both: Combine speed with power
When to use thinking mode?

Enable thinking mode for:

  • Complex analytical questions
  • In-depth text analysis
  • Multi-step reasoning
  • Important decisions

Claude Sonnet 4.5​

Introduction Claude Sonnet 4.5 is Anthropic's latest advanced AI language model. It combines excellent reasoning capabilities with a large context window and strong ethical principles.

Key Features

  • Size: 200+ billion parameters
  • Context window: 200,000 tokens
  • Multilingual support for 100+ languages
  • Advanced code generation capabilities
  • Superior logical reasoning
  • Accurate fact-checking
  • Extensive capabilities for long documents
  • Improved bias mitigation

Applications

  • Code assistance and development
  • Long document analysis
  • Chatbots and virtual assistants
  • High-quality language translation
  • Business support
  • Creative content creation

Limitations

  • Knowledge up to training data cutoff
  • Sometimes cautious answers on sensitive topics

GPT 5.2​

Introduction GPT 5.2 is the latest generation of OpenAI's language models, with significant improvements in reasoning, creativity, and reliability.

Key Features

  • Improved multimodal understanding
  • Advanced logical reasoning
  • Greater accuracy for factual questions
  • Improved code generation
  • Faster response times
  • Native tool use (web search, code execution)

Applications

  • Advanced chatbots
  • Personal assistance
  • Creative content generation
  • Code generation and debugging
  • Strategic analysis
  • Data analysis

Limitations

  • Knowledge up to training data cutoff (use web search for current info)
  • May need multiple attempts for complex tasks

Gemini 3 Pro​

Introduction Gemini 3 Pro is Google's most advanced multimodal AI model, optimized for both text and image.

Key Features

  • Native multimodal processing
  • Very large context window (1M+ tokens)
  • Fast response time
  • Superior understanding of images and documents
  • Advanced reasoning
  • Strong multilingual support

Applications

  • Research and analysis
  • Image and document analysis
  • Contextual content generation
  • Complex task processing
  • Creative writing
  • Technical support

Limitations

  • May be less detailed than specialized models for very specific tasks

Smaller / Fast Models​

In addition to the large models, AI-Public also provides access to faster, cheaper alternatives:

ModelProviderIdeal for
GPT 4.1 MiniOpenAIQuick questions, simple tasks
Claude Haiku 3.5AnthropicShort texts, translations
Gemini 3 FlashGoogleFast responses, basic tasks

These models are perfect as a "fast model" in combination with a thinking model.