Major BotHub Telegram Bot Update: Video Generation, Automatic Model Selection, and More

We've rolled out a major update to our Telegram bot, bringing new AI generation tools, automatic model selection, more customization options, and improvements to the chat experience.You can now create videos and images directly in Telegram, let us select the right AI model for your task, and configure individual chats to match your workflow.Here's what's new.

New Tools

Video Generation in Telegram

We've added video generation to our Telegram bot, allowing you to create videos from text prompts or images without switching to another application.Available models include:

Veo 3.1: Standard, Fast, and Lite versions.
Hailuo 2.3.
Runway: Gen-4.5 and Gen-4 Turbo.

We display the estimated generation cost before you start, so you can check how many Caps your request will use.Please note: Video generation typically consumes significantly more Caps than text-based requests.

Image Generation

We've expanded the bot's image generation capabilities.You can now select an image generation model independently for each chat, making it easier to work on different creative and professional tasks without repeatedly changing your settings.

New Features

Automatic AI Model Selection

We've introduced automatic model selection to help you find the right AI model without comparing dozens of options manually.The bot analyzes your request and selects a suitable model based on the task, including:

Image analysis: Understanding and analyzing uploaded images.
File processing: Reading and working with documents.
Web search: Finding information online.
General questions: Handling everyday conversations and requests.

Automatic model selection is enabled by default for new users.If you've used the bot before, you can enable it in the settings by choosing Model Selection: Auto.You can switch back to manual model selection whenever you prefer.Availability: Automatic model selection is also available on the free plan.

Improve Answer

We've added an Improve Answer button to help you get a more detailed or higher-quality response.If you're not satisfied with an answer, simply tap the button below the message. We'll use a more capable AI model to generate an improved version.This feature is available when automatic model selection is enabled.

Real-Time Response Streaming

The bot now displays text as it's being generated instead of waiting until the entire response is complete.You can start reading the answer immediately while the model continues processing your request.

Expanded Settings Menu

We've introduced a new settings menu that gives you more control over how the bot works.

1. Presets

Configure chats using ready-made or custom presets:

Popular predefined chat roles.
Preset categories.
Custom configurations.
Company presets.
Favorites.

2. Tools

Enable or disable individual tools depending on your tasks:

Web search.
Image generation.
File creation, including DOCX, XLSX, and code files.
Scientific paper search.
Legal information search.
Link analysis.
GitHub integration.

3. Display and Preferences

Customize how the bot presents information:

Read responses aloud.
Show model reasoning.
Enable or disable animated emojis.
Display usage costs in Caps or RUB.
Adjust response streaming settings.

4. Reasoning Effort

For models that support configurable reasoning, you can choose between three levels:

Minimum: Prioritizes faster responses.
Medium: Balances response speed and analytical depth.
Maximum: Allocates more reasoning effort to complex tasks.

These options let you adjust the bot's behavior to suit your workflow and the complexity of your requests.

Improvements

Clearer Response Formatting

We've improved how the bot displays different types of content to make AI-generated answers easier to read.

Tables

Tables now use a clearer, more structured layout, making it easier to review and compare information.

LaTeX Formulas

We've improved mathematical formula rendering. LaTeX expressions now appear in a readable format instead of displaying raw markup.

Model Reasoning

You can expand the model's reasoning above the final response when this information is available.

Corporate Account Balance

We've improved balance visibility for employees using BotHub through corporate accounts.The /start command now displays the employee's available corporate balance instead of their personal balance.This matches the balance shown on the BotHub website, helping corporate users keep track of their available Caps directly in Telegram.

Get Started

All these features are available in our updated Telegram bot (@bothub_bot)!Try the new generation tools, enable automatic model selection, and configure the bot for your everyday tasks.