Changelog - v3.1.5
✨ New
AI Models
The focus of this release is support for the latest AI models from OpenAI, Anthropic, Mistral, and Meta.
Based on these new models, we've also updated our question credit system to be more accurate and fair. More specifically, we've scaled the question credits to be more in line with the cost of the models. You now have twice the question credits you used to 🎉, which means you get 10,000 questions per month in the Pro plan with GPT-4o mini (the most cost-efficient model available).
The full model list is now as follows:
Provider | Model | Credits/Question | Intelligence | Speed | Cost Efficiency | Hosted In | Available In |
---|---|---|---|---|---|---|---|
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.
We are especially excited about Claude 3.5 Sonnet. This marks a significant milestone as it's the first top-tier AI model we can offer with complete EU data residency to all our users, without requiring Azure OpenAI integration in the Enterprise plan. Please do enjoy the end-to-end data residency within the European Union, without sacrificing on quality.
Analytics
We've added two powerful new metrics to our analytics dashboard:
- Average Handling Time (AHT): This metric reveals the duration between a user's initial message and the AI's final response in a chat session.
- CTA Click Rate: This shows the percentage of users who engage with Call-to-Action buttons during chat sessions where they've interacted with the AI at least once.
To give you more control over your data analysis, we've also introduced filtering options:
- Chatbot Filter: Analyze data for specific chatbots, allowing for targeted improvements.
- Date Range Selector: Easily compare data across different time periods by selecting specific months to view.
🛠️ Improved
- Google Drive: Improved stability and notifications for the Google Drive integration.
- Concurrent Chatbot Editing: Improved handling of simultaneous chatbot edits by multiple users. Now, when you save your changes on a chatbot, we will only apply your changes to the chatbot and not overwrite the changes made by other users. However, we still prioritize the latest changes, which means if another user and you edit the same property of a chatbot, the latest change will be applied to that property.
🐛 Fixed
- Fixed several bugs and issues.