Inside the Hatz AI platform, we’re already at over 50 models, and that number keeps growing. That’s why the focus shouldn’t be on memorizing which models are good at certain tasks, but on recognizing when you should be using them.
There’s already a breakdown in the Hatz platform itself that shows what each model is good for — whether it’s advanced reasoning, tool calling, image generation, or speed. Read more here. As a general rule of thumb, we can guide customers:
Chat and text models → what people are using the most today, very powerful for writing and conversation.
Reasoning models → more advanced, ideal for problem-solving and decision-making.
Models with tool calling → critical for multi-step workflows, data analysis, or integrations.
Vision models → good at seeing and interpreting images.
Image generation models → good at creating visuals and designs quickly.
Regarding some Tools...
Google Search: Quick fact-checking and immediate answers
Perplexity: Current news, trends, and event analysis
Firecrawl: Comprehensive website data collection and extensive analysis
Tavily & Exa: Alternative AI search engines for varied results
The key message is: just start. Experiment with different LLMs, tools, and integrations. Get a feel for the differences, and then scale from there. Once you move beyond simple chat into workflows and agents, it’s a whole different context — because then you’re not just asking ‘what tool is good for writing?’ You’re thinking about how AI can run multi-step processes automatically for your business.”