Most popular AI assistant. GPT-4 powers the paid version (ChatGPT Plus). Features: web browsing, file uploads, image generation (DALL-E), code interpreter.
Best for: General conversation, writing assistance, brainstorming
Limitation: Knowledge cutoff (January 2024 for GPT-4 Turbo)
Pricing: Free (GPT-3.5) / $20/month (GPT-4)
📌 Claude (Anthropic)
Known for strong reasoning, safety, and large context window (200k tokens). Excellent for analyzing long documents.
Best for: Long document analysis, complex reasoning, coding
Differentiator: "Constitutional AI" – designed to be helpful and harmless
Pricing: Free (Claude 3 Haiku) / subscription for Opus
📌 Gemini (Google)
Multimodal by default – supports text, images, audio, and video. Integrated with Google Workspace.
Best for: YouTube summaries, Google Drive integration, multimodal tasks
Differentiator: 1M token context window (can process entire books)
Pricing: Free / $19.99/month (Gemini Advanced)
📌 Perplexity AI
AI-powered search engine with citations. Combines LLM with real-time web search.
Best for: Research, fact-checking, current events
Differentiator: Provides sources for every answer
Pricing: Free / $20/month (Pro)
💻 AI Coding Tools
📌 GitHub Copilot
AI pair programmer integrated into VS Code, JetBrains, and other IDEs. Autocompletes code based on context and comments.
Best for: Autocompletion, boilerplate code, learning new APIs
Works with: All major languages (Python, JavaScript, TypeScript, Go, Rust, etc.)
Pricing: $10/month or $100/year
📌 Cursor
AI-first IDE built on VS Code. Features: Cmd+K for inline editing, chat with codebase, AI terminal commands.
Best for: Full AI-powered development environment
Differentiator: Can reference entire codebase, not just open files
Pricing: Free / $20/month
📌 Cline (prev. Claude Dev)
VS Code extension that handles file editing, terminal commands, and browser automation.
Best for: Autonomous code changes, file operations
Differentiator: Can create and edit files, run terminal commands, use browser
Pricing: Free (bring your own API key)
📌 Continue
Open-source VS Code extension for AI code assistance. Works with local models or cloud APIs.
Best for: Privacy-focused developers
Differentiator: Can run models locally (Ollama, Llama.cpp)
Pricing: Free
🏗️ Development Frameworks
📌 LangChain
Most popular framework for building LLM applications. Provides chains, agents, memory, and integrations with 100+ tools.
Best for: Complex LLM applications with multiple steps and tools
Languages: Python, TypeScript/JavaScript
Pricing: Free (open-source)
📌 LlamaIndex
Specialized framework for RAG (retrieval-augmented generation) applications. Excellent for document indexing and querying.
Best for: RAG, document Q&A, data connectors
Differentiator: Many data loaders (PDFs, databases, APIs)
Pricing: Free (open-source)
📌 Semantic Kernel (Microsoft)
Microsoft's lightweight framework for AI orchestration. Integrates well with Azure services.
Best for: .NET shops and Azure users
Languages: C#, Python, Java
Pricing: Free
📌 Ollama
Run open-source LLMs locally (Llama 3, Mistral, Phi-3) on your own hardware.
Best for: Privacy, offline use, no API costs
Requirements: Local GPU (or CPU for smaller models)
Pricing: Free
🔄 Workflow & Agent Tools
📌 n8n (AI-native workflow)
No-code automation platform with AI nodes. Connect LLMs to APIs, databases, and services.
Best for: Building AI automations without code
Pricing: Free self-hosted / paid cloud
📌 Flowise
Drag-and-drop UI for building LangChain applications. Great for prototyping.
Best for: Visual prototyping of LLM flows
Pricing: Free
📌 Vector Databases (Quick reference)
Pinecone – Managed, fast, pay-as-you-go
Chroma – Lightweight, embedded, great for prototypes
Qdrant – Open-source, written in Rust, very fast
Weaviate – Open-source with hybrid search
pgvector – Vector search inside PostgreSQL
📊 Tools Comparison Table
Tool
Category
Cost
Best for
ChatGPT
Chatbot
Free/$20
General conversation
Claude
Chatbot
Free
Long documents, reasoning
GitHub Copilot
Coding
$10/mo
Autocomplete
Cursor
Coding
Free/$20
AI-first IDE
LangChain
Framework
Free
Complex LLM apps
LlamaIndex
Framework
Free
RAG applications
Ollama
Local
Free
Run models locally
n8n
Workflow
Free/paid
AI automations
✍️ Exercises
Exercise 1.05.1 – Choose the right tool
For each scenario, choose the best tool from the list above:
You need an AI to analyze a 500-page legal document.
You're building an internal RAG system for company documents and want an open-source framework specialized for this.
You want autocomplete in your VS Code while writing Python.
You need to process customer documents on-premise for privacy reasons (no cloud APIs).
Answers:
1. Claude (200k context window, good for long documents)
2. LlamaIndex (specialized for RAG)
3. GitHub Copilot or Cursor
4. Ollama (run local open-source models)
Exercise 1.05.2 – Tool research
Pick one tool from the list above, visit its website, and write a one-paragraph summary of its key features and pricing.
Example for Cursor:
Cursor is an AI-first code editor built on VS Code. Key features include: Cmd + K for inline editing (write or modify code with natural language), chat with your entire codebase, AI-predictions for what you'll type next, and integrated AI terminal commands. Pricing: Free tier includes 2 weeks of premium features, then $20/month for Pro (unlimited AI commands, chat with codebase). Cursor uses a mix of GPT-4, Claude 3, and their own models.
Exercise 1.05.3 – Framework exploration
Go to the LangChain documentation website and find one example of a "chain" (e.g., LLMChain). Describe what it does in one sentence.
Example:
An LLMChain is the simplest chain that combines a prompt template with an LLM – you give it user input, it formats the prompt, sends it to the LLM, and returns the response. For example, a chain that summarizes user-provided text.
📌 Tool Selection Tips
Start with general-purpose chatbots (ChatGPT, Claude) for exploration
For coding, try GitHub Copilot or Cursor – they save significant time
For RAG applications, LangChain + vector DB (e.g., Pinecone) is a common stack
For privacy/offline, use Ollama with local models (Llama 3, Mistral)