Poll: Which is your favorite AI model in Gladys?

[pierre-gilles]

Hello everyone!

I recently added the ability to select an AI model in Gladys, and I’d like to get your feedback on the topic: which model do you find the most intelligent for home automation use (understanding requests, relevance of responses, etc.)?

If you’re not using Gladys Plus, there’s a free month without a credit card, so don’t hesitate, it’s made for you

Feel free to test multiple models and vote for your favorites (max 3 choices). Your feedback will help me decide which ones to highlight by default in Gladys!

• Mistral 3.2 (€)
• Gemma 4 (€)
• Qwen 3.6 35B (€€)
• Pixtral 12B (€)
• Gemma 3 27B (€)
• Holo2 30B (€)
• Qwen 3.5 397B (€€€)
• Llama 3.3 70B (€€)
• Qwen 3 235B (€€€)

Thanks again to everyone who will test

[Will_71]

For my part, I don’t use them (yet).

[mutmut]

I tried in Auto but we don’t know which one is taken: the cheapest? the fastest?
The results weren’t great for the same question asked, even when changing with the top three:

After that, when I changed, I got an error that the AI was unavailable (my credits were fine).

And when we have the weekly summary, well I have the right info (for the pool).

I’ll continue to test and try to document a bit more, and not blame the AI when it doesn’t understand anything

[pierre-gilles]

@mutmut

I tested in Auto, but we don’t know which one is taken: the cheapest? the fastest?

Mistral 3.2 Small!

But precisely, I have feedback that for some people it isn’t so intelligent, surely it gets a bit stupid as the context increases. I’m looking for a new default model

[Chris75]

From my side, I’ve conducted several real-world tests on my current setup, which includes lights, shutters, outlets, switches, and cameras. The one that performs the best is Qwen 3.6 35B (60% success rate compared to 10% for Mistral), but it’s far from perfect compared to Alexa+ or Google, which understand over 95%! But this is just my opinion and doesn’t commit anyone else!