AI Model
Dashworks supports the following LLMs. You can choose the LLM for each question from the toolbar in Dashworks' input box, set the default for your use here, and set the default for the entire workspace as an Admin here.
Here is a general guideline on when to use:
GPT-4o: Default model, best overall balance of accuracy, speed, instruction following, response length, and response tone. Use this for most tasks.
Claude Sonnet 3.5: Use this when drafting content (ex. emails, blog post, social media posts), writing or debugging code, or for long prompts with complex instructions.
Gemini Pro: Use this if you want to minimize error rate in factual questions. It will more frequently respond with "I don't know" than other models if it does not have high confidence in its response.
Model | Accuracy | Speed | Complex Instructions | Natural Sounding |
---|---|---|---|---|
GPT-4o | High | High | High | Medium |
GPT-4-Turbo | Medium | Medium | High | Low |
Claude Sonnet 3.5 | High | High | Medium | High |
Claude Opus | High | Low | High | High |
Gemini Pro 1.5 | High | Medium | High | Medium |
Meta Llama 3 | Medium | High | Low | High |
Last updated