Appearance

Advanced Features ​

Discover Beloga's powerful additional features that enhance your productivity and flexibility.

Analytics Dashboard ​

Usage Insights ​

Available Metrics

Track your usage patterns with:

  • Search patterns and frequency
  • Most accessed sources
  • Popular search topics
  • Team collaboration stats
  • AI model usage

Activity Tracking ​

Track Your Progress

Monitor your:

  • Daily search volume
  • Knowledge base growth
  • Collaboration frequency
  • Resource utilization

AI Model Selection ​

Beloga offers a wide range of AI models to suit different needs and use cases. Models are organized by provider and capability level.

Default Model ​

BLG-1.0

Beloga's flagship model that automatically routes your queries to the best model for your specific needs. Available to all users.

OpenAI Models ​

  • GPT-4o-mini: Optimized version of GPT-4o for faster responses
  • GPT-4o: Enhanced GPT-4 model with improved capabilities (Pro)
  • GPT-4 turbo: Latest version with improved speed and capabilities (Pro, 30 requests/day)
  • o1-mini: New model designed for reasoning and problem-solving (Pro, 10 requests/day)

Anthropic Models ​

  • Haiku-3.5: Fast and efficient version of Claude 3
  • Sonnet-3.5: Balanced version of Claude 3 with strong capabilities (Pro)
  • Opus-3: Most powerful version of Claude 3 with advanced reasoning (Pro)

Google Models ​

  • Gemini-1.5-Flash: Fast and efficient version of Google's Gemini 1.5
  • Gemini-1.5-Pro: Pro version with enhanced capabilities (Pro)
  • Gemini-2.0-Flash-exp: Experimental Gemini 2.0 Flash model (Pro, Beta)
  • Gemini-2.0-Flash-Thinking-exp-1219: Latest experimental model with reasoning capabilities (Pro, Beta, New)
  • LearnLM-1.5-pro: Experimental learning-focused language model (Pro, Beta)
  • Experimental Models: Multiple beta versions available (Gemini-exp-1114, 1121, 1206)

Other Providers ​

Meta ​

  • Llama-3.3-70B: Latest 70B parameter model with improved instruction following (Pro)

NVIDIA ​

  • Llama-3.1-Nemotron: 70B parameter model based on Llama architecture (Pro)

X.AI ​

  • Grok-2.0: Latest model known for witty responses (Pro, New)

Liquid ​

  • LFM-40b: 40B parameter foundation model (Pro)

Alibaba ​

  • Qwen-2.5-coder: Specialized coding model (Pro)
  • QwQ-32b-preview: 32B parameter model designed for reasoning (Pro, Beta)

DeepSeek ​

  • DeepSeek-2.5: Latest chat model optimized for dialogue (Pro)

Amazon ​

  • Nova-Lite-V1: Lightweight model optimized for efficiency (Pro)

Model Access ​

Access Levels

  • Free Users: Access to BLG-1.0, Haiku-3.5, and Gemini-1.5-Flash
  • Pro Users: Access to all models, including experimental and specialized ones

Model Selection

Beloga's default BLG-1.0 model automatically selects the best model for your query. Manual model selection allows you to:

  • Choose specific models for specialized tasks
  • Test experimental models
  • Control response speed vs. capability tradeoffs

Model Settings ​

Customization Options

Adjust your AI experience:

  • Select default model
  • Set fallback options
  • Configure context length
  • Adjust response style

Use Cases ​

Model Selection Guide
  1. GPT-4: Complex reasoning and analysis
  2. Claude: Long-form content and documentation
  3. Gemini: Real-time data and code analysis
  4. Open Source: Privacy-sensitive operations

Freestyle Mode ​

Direct AI Access ​

Settings

Customize your AI interaction:

  • Choose your preferred model
  • Set custom instructions
  • Enable conversation memory
  • Configure response format

Features ​

Experience pure AI interaction:

  • No source integration
  • Direct AI conversations
  • Custom instructions
  • Conversation memory
  • Flexible interactions

WARNING

Freestyle mode bypasses source verification - use with awareness of potential hallucinations

Content Capture ​

Chrome Extension ​

Extension Features

Save content easily with:

  • One-click saving
  • Smart categorization
  • Auto-summarization
  • Context preservation

Capture Methods ​

Capture Options

Save content your way:

  • Full page capture
  • Text selection
  • Key points extraction
  • Auto-generated summaries

Supported Captures ​

Save various content types:

  • Web pages
  • PDF documents
  • Images with text
  • Code snippets
  • Social media posts

Integration ​

Seamless Workflow

  1. Capture content while browsing
  2. Auto-organize into your library
  3. Instant searchability
  4. Share with team members

Best Practices ​

Model Selection ​

Tips for Model Choice
  1. Consider task complexity
  2. Evaluate context length needs
  3. Balance speed vs accuracy
  4. Account for cost considerations

Content Organization ​

Organization Structure

Organize your captures into:

  • Web Content
  • Documents
  • Images
  • Code References

Analytics Usage ​

Optimization Strategies

  1. Review search patterns
  2. Identify knowledge gaps
  3. Optimize team collaboration
  4. Adjust AI model selection