🎥 Resources & Setup

Tutorials, guides, and troubleshooting in one place.

Watch the latest walkthroughs, configure WebGPU/Ollama/OpenAI, and solve common setup questions.

Watch Tutorials Open Setup Guide

Learn SotaSearch.ai

Watch our comprehensive video tutorials to master every feature of SotaSearch.ai

Featured: SotaSearch.ai Tutorial

Discover the latest features and capabilities of SotaSearch.ai in this comprehensive tutorial that showcases the power of privacy-first, AI-driven research assistance.

Featured Latest Features AI Research Privacy First

Complete Research Workflow Tutorial

Master the full SotaSearch.ai research process: select papers, generate state-of-the-art reviews, create citation graphs, and generate novel paper abstracts using AI-powered insights.

Paper Selection Literature Review Citation Graphs Paper Abstracts AI Chat

Popup Menu & Library Management

Learn how to use SotaSearch's popup menu for chatting with webpages, summarizing content, and managing your research library with PDF uploads and metadata extraction.

Webpage Chat PDF Processing Library Management Smart Summarization

Configuration & Setup Guide

Complete setup tutorial for configuring SotaSearch.ai with different AI providers, API keys, and local models for optimal performance and privacy.

AI Configuration API Setup Local Models Privacy Settings

Ready to Master SotaSearch.ai?

Watch these tutorials and transform your research workflow with AI-powered insights and complete privacy.

Join Waitlist Watch Demo

Choose your setup path

WebGPU, local Ollama, OpenAI, and troubleshooting resources in one place.

Enable WebGPU (Chrome)

WebGPU lets SotaSearch.ai run WebLLM fully in your browser without external servers.

1

Enable WebGPU Flag

Navigate to chrome://flags/#enable-unsafe-webgpu and set it to "Enabled".

2

Restart Chrome

Close all Chrome windows completely and reopen to apply the flag.

3

Check GPU Support

Visit chrome://gpu and confirm WebGPU shows as Enabled.

4

Load WebLLM

Open the SotaSearch.ai extension and select a WebLLM model (Qwen2.5 3B, Llama 3 etc.).

Benefits of WebGPU:

  • No installs: Everything runs in the browser.
  • Privacy: Data never leaves the page.
  • Speed: Perfect for quick summaries or lightweight reviews.

Setup Ollama for Local AI

Ollama allows you to run AI models locally on your machine with full privacy and control.

1

Install Ollama

Download and install Ollama from ollama.ai

2

Set Extension Origin

Configure Ollama to accept connections from the SotaSearch.ai extension:

Windows PowerShell 
setx OLLAMA_ORIGINS "chrome-extension://*,extension://*,http://localhost:11434"
                                                        ollama serve && ollama serve
macOS/Linux Terminal 
export OLLAMA_ORIGINS="chrome-extension://*,extension://*,http://localhost:11434" && ollama serve
3

Download AI Models

Download your preferred AI models using Ollama:

Example Models 
ollama pull llama3.1:8b
ollama pull mistral:7b
ollama pull codellama:7b
4

Test Connection

Verify that Ollama is running and accessible from the extension. You should see "Ollama Connected" in the extension settings.

Benefits of Using Ollama:

  • Full Privacy: AI models run entirely on your machine
  • No Internet Required: Works offline once models are downloaded
  • Custom Models: Use any model compatible with Ollama
  • Better Performance: Faster response times with local processing

Setup OpenAI API Integration

Integrate OpenAI's powerful AI models like GPT-4 and GPT-3.5 through their API for enhanced research capabilities.

1

Create OpenAI Account

Visit OpenAI Platform and create an account to access their API services.

2

Generate API Key

Navigate to the API Keys section in your dashboard and click "Create new secret key" to generate your API key.

🔑

Security Note

Copy and save your API key immediately - it won't be visible again after closing the dialog.

3

Configure in Extension

Add your OpenAI API key to the SotaSearch.ai extension settings:

Extension Settings 
1. Open SotaSearch.ai extension
2. Go to Settings → AI Providers
3. Select OpenAI
4. Enter your API key
5. Choose model (GPT-4, GPT-3.5-turbo, etc.)
6. Set usage limits if needed
4

Test Connection

Verify your OpenAI integration by asking a research question in the extension. You should see responses from the selected OpenAI model.

Benefits of Using OpenAI:

  • Advanced Models: Access to GPT-4, GPT-3.5-turbo, and other cutting-edge AI models
  • High Quality: Superior reasoning and analysis capabilities for complex research tasks
  • Reliability: Enterprise-grade API with high uptime and performance
  • Continuous Updates: Always access to the latest model improvements and features
💰

Pricing Information

OpenAI API usage is charged per token. Check OpenAI's pricing page for current rates and set usage limits in your account to control costs.

Common WebGPU Issues & Solutions

WebGPU is experimental technology. Here are solutions to common problems:

❌

WebGPU Not Available

Problem: "WebGPU is not supported" error

Solution:

  • Update Chrome to version 113+ (Windows/macOS) or 121+ (Android)
  • Enable WebGPU flag: chrome://flags/#enable-unsafe-webgpu
  • Restart Chrome completely
🐌

Slow Performance

Problem: Very slow AI processing

Solution:

  • Ensure hardware acceleration is enabled
  • Update GPU drivers to latest version
  • Close other GPU-intensive applications
  • Consider using Ollama for better performance
💥

Browser Crashes

Problem: Browser crashes during AI processing

Solution:

  • Reduce model size (use smaller WebLLM models)
  • Increase browser memory allocation
  • Use Ollama instead of WebGPU
  • Report issues to our support team
🔧

Alternative Solutions

If WebGPU doesn't work:

  • Ollama: Local models with better performance
  • OpenAI API: Cloud-based with high reliability
  • OpenRouter: Multiple AI providers
💡

Performance Tips

  • Model Selection: Start with smaller models (7B parameters) for faster processing
  • Hardware: Dedicated GPU recommended for optimal performance
  • Browser: Chrome 113+ with latest updates provides best WebGPU support
  • Memory: 8GB+ RAM recommended for smooth operation

Ready to Get Started?

Choose your preferred AI setup method and start using SotaSearch.ai with powerful AI models.

Join Waitlist Download Ollama Get OpenAI API