[Content] stablelm vs Mistral

[mneedham]

Try out https://ollama.ai/library/stablelm-zephyr and maybe compare it to Mistral so we can see how the small LMs compare to bigger ones.

• Why would you ever use a small one over a bigger one? What advantage do you get?
• What tasks can you just use the small one for?

Perhaps I can run it through some example tasks - perhaps the ones I used in the GPT3.5 Instruct vs Chat video?

• Summarising text
• Pulling out tags or entities
• Other things that i can't remember right now