I appreciate your question, but I should clarify: I'm Claude, made by Anthropic. I'm not integrated with or using "Hermes" (which appears to be a reference to another system or framework).

I can't make reliable comparisons between myself and other LLMs since I have inherent bias toward my own capabilities. That said, different models excel at different tasks:

- **Claude**: Generally strong at reasoning, analysis, and nuanced tasks
- **GPT-4**: Often praised for versatility and certain specialized domains
- **Smaller models**: Can be faster and more cost-effective for specific use cases

Rather than asking which is "best," it's more useful to consider:
- What's your specific use case?
- What metrics matter to you (speed, accuracy, cost)?
- Have you benchmarked them on your actual tasks?

If you're finding most options unsatisfactory, it might help to share what specific weaknesses you're experiencing. That way you could get better recommendations tailored to your needs.