Used model-agnostic examples in quickstart / first crew / first flow guides

[markmcd]

Updated prereqs and setup steps to point to the now-more-model-agnostic LLM setup guide, and updated the relevant text to go with it.

Hopefully this encourages folks to explore the available models.

[joaomdmoura]

Disclaimer: This review was made by a crew of AI Agents.

Code Review Comment for PR #2773

Overview

This PR effectively transforms the documentation by making it model-agnostic rather than being exclusively OpenAI-specific. This update significantly enhances flexibility for users across different LLM providers, broadening the potential user base and improving usability within the CrewAI framework.

File-by-File Suggestions with Examples

1. docs/guides/crews/first-crew.mdx

Positive Changes:

• Inclusion of a reference to the LLM setup guide, enhancing directions for users.
• Transition to provider-agnostic agent configurations, fostering inclusivity.

Suggestions for Improvement:

• Environment Variables Section:

  # Current
  SERPER_API_KEY=your_serper_api_key
  # Add your provider's API key here too.
  
  # Recommended
  SERPER_API_KEY=your_serper_api_key
  # LLM Provider API Keys (uncomment the one you're using)
  # OPENAI_API_KEY=your_openai_api_key
  # ANTHROPIC_API_KEY=your_anthropic_api_key
  # GOOGLE_API_KEY=your_google_api_key
  

• Agent Configuration Example:

  # Current
  llm: provider/model-id  # e.g. openai/gpt-4o, google/gemini-2.0-flash
  
  # Recommended
  llm: provider/model-id  
  # Supported formats:
  # - OpenAI:     openai/gpt-4
  # - Anthropic:  anthropic/claude-3
  # - Google:     google/gemini-pro
  # See full list in documentation at /concepts/llms
  

2. docs/guides/flows/first-flow.mdx

Positive Changes:

• Updated multi-provider environment variable examples, reinforcing the document's inclusivity.

Suggestions for Improvement:

• LLM Configuration Example:

  # Current
  llm = LLM(
      model="model-id-here",  # gpt-4o, gemini-2.0-flash
  
  # Recommended
  llm = LLM(
      model="provider/model-id",  
      # Supported formats:
      # - OpenAI:     openai/gpt-4
      # - Anthropic:  anthropic/claude-3
  )
  

3. docs/quickstart.mdx

Positive Changes:

• Removal of specific references to OpenAI allows for broader usability.

Suggestions for Improvement:

• Agent Configuration Example:

  # Current
      llm: provider/model-id  # Add your choice of model here
  
  # Recommended
      llm: provider/model-id
      # Configure with your chosen provider:
      # OpenAI:    openai/gpt-4
  

General Recommendations

1. Documentation Consistency:

   • Ensure uniform formatting for code examples across all files.
   • Add version compatibility notes for each LLM provider to ease user navigation.

2. Configuration Examples:

   • Incorporate links to specific setup guides for major LLM providers for quick access.
   • Outline minimum required versions to avoid compatibility issues.

3. Error Handling:

   • Implement troubleshooting sections addressing common setup problems.
   • Validate API keys with examples to quickly assist users encountering issues.

4. Best Practices:

   • Include model selection guidance tailored to various contexts and user requirements, especially concerning cost and operational performance.

Historical Context & Related PR Insights

Recent trends in related pull requests indicate a collective move towards enhancing documentation for multidimensional user bases. This shift signifies not only a growing user base but a forward-thinking approach to documentation structure that anticipates the evolving landscape of LLM technologies and user preferences.

By maintaining an inclusive stance and refining your documentation through these suggested improvements, you will align well with best practices and user expectations in today’s environment. Let's continue to focus on enhancing user experiences across different setups.

Thank you for your efforts on this PR!