Changelog - v3.8.0
Introducing GPT-5 nano and mini variants for flexible AI model selection and cost optimization.
Last edited: 9/27/2025We believe in creating a tool that speaks for itself, powered and shaped by your support and feedback. So please don't hesitate to reach out to us with your thoughts and suggestions.
✨ New
GPT-5 Variants
The focus of this release is introducing GPT-5 nano and mini variants, providing more flexible AI model options for different use cases and budgets.
We've added comprehensive GPT-5 variant support:
- GPT-5 Mini: Optimized for speed and cost-efficiency with 3/5 intelligence rating, perfect for high-volume applications requiring fast responses at 8 credits per question.
- GPT-5 Nano: Ultra-fast and most cost-effective GPT-5 variant with 2/5 intelligence rating, ideal for simple tasks and rapid prototyping at 5 credits per question.
- Flexible Model Selection: Choose the right GPT-5 variant based on your specific needs - maximum intelligence (GPT-5), balanced performance (GPT-5 Mini), or speed and cost (GPT-5 Nano).
- Enhanced Cost Optimization: More granular pricing options allowing users to optimize their AI spend based on task complexity and response requirements.
AI Model Performance
Provider | Model | Credits/Question | Intelligence | Speed | Cost Efficiency | Hosted In | Available In |
---|---|---|---|---|---|---|---|
OpenAI | GPT-5 | 15 | ◼︎◼︎◼︎◼︎◼︎ | ◼︎◼︎◼︎◻︎◻︎ | ◼︎◻︎◻︎◻︎◻︎ | 🇺🇸* | Pro |
OpenAI | GPT-5 Mini | 8 | ◼︎◼︎◼︎◼︎◻︎ | ◼︎◼︎◼︎◼︎◻︎ | ◼︎◼︎◼︎◼︎◻︎ | 🇺🇸* | Pro |
OpenAI | GPT-5 Nano | 5 | ◼︎◼︎◼︎◻︎◻︎ | ◼︎◼︎◼︎◼︎◼︎ | ◼︎◼︎◼︎◼︎◼︎ | 🇺🇸* | Pro |
OpenAI | GPT-4o mini | 1 | ◼︎◼︎◼︎◼︎◻︎ | ◼︎◼︎◼︎◻︎◻︎ | ◼︎◼︎◼︎◼︎◼︎ | 🇺🇸* | Pro, Essential |
OpenAI | GPT-4 Turbo | 25 | ◼︎◼︎◼︎◼︎◻︎ | ◼︎◻︎◻︎◻︎◻︎ | ◼︎◻︎◻︎◻︎◻︎ | 🇺🇸* | Pro |
OpenAI | GPT-4o | 12 | ◼︎◼︎◼︎◼︎◼︎ | ◼︎◼︎◻︎◻︎◻︎ | ◼︎◼︎◼︎◻︎◻︎ | 🇺🇸* | Pro |
Mistral | Mistral Large 2 | 8 | ◼︎◼︎◼︎◼︎◻︎ | ◼︎◻︎◻︎◻︎◻︎ | ◼︎◼︎◼︎◻︎◻︎ | 🇪🇺 | Pro |
Anthropic | Claude 3 Haiku | 1 | ◼︎◻︎◻︎◻︎◻︎ | ◼︎◼︎◼︎◻︎◻︎ | ◼︎◼︎◼︎◼︎◼︎ | 🇪🇺 | Pro, Essential |
Anthropic | Claude 3.5 Sonnet | 12 | ◼︎◼︎◼︎◼︎◼︎ | ◼︎◼︎◻︎◻︎◻︎ | ◼︎◼︎◼︎◻︎◻︎ | 🇪🇺 | Pro |
Meta | Llama 3.1 | 1 | ◼︎◼︎◼︎◻︎◻︎ | ◼︎◼︎◼︎◼︎◼︎ | ◼︎◼︎◼︎◼︎◼︎ | 🇺🇸 | Pro, Essential |
* Hosted in EU when using Azure OpenAI integration in the Enterprise plan, otherwise hosted in US.
🛠️ Improved
- Enhanced Model Selection: Added GPT-5 Mini and Nano variants providing more granular options for different use cases and budgets.
- Better Cost Optimization: Users can now choose between GPT-5 (maximum intelligence), GPT-5 Mini (balanced performance), or GPT-5 Nano (speed and cost) based on their specific needs.
- Expanded Pro Plan Value: Pro plan users now have access to three GPT-5 variants - 666 GPT-5 questions, 1,250 GPT-5 Mini questions, or 2,000 GPT-5 Nano questions per month.
🐛 Fixed
- Fixed several bugs and improved overall system performance.
For detailed information about DocuChat features, please visit our documentation.