Set aside benchmarks, if you had to choose one to use instead of ChatGPT for the next 6 months, which one would you pick? Recently, I’ve been experiencing some extreme slow down and poor answers on GPT so I’m going to run a local backup for the time being to assist when GPT4 is down. I’m leaning towards Mistral. I can be convinced to test some others, though.
You must log in or register to comment.
I keep trying new models, and I keep going back to Dolphin-Mistral-2.2.1. There is something about the quality of the interactions that is different from the other models, and is, I don’t know, unexplainably better. I cannot identify why this remains in my mind the best model of all models I’ve tested, clear up to 33b (the largest my pitiful machine will load), but I continue to think this. Now, I haven’t tested every model, so my opinion is completely anecdotal. Dolphin just kicks it, though. It just does such a good job at almost everything I throw at it. I won’t say it doesn’t foul up here and there, but it still blows the other small models out of the water as far as I’m concerned.
There is an upcoming NeuralHermes-2.5-Mistral-70B, chances are it will also have vision version as well. Looking at really impressive performance of 7B version. I think 70B will set new benchmarks in OSS AI world. But, there are plenty of other models as well. You should choose according to your use case.
The Best? Id think a goliath 120B finetune like Tess XL
Lol 😂 I’d need an upgrade
You can always test it out using runpod for around ~2 dollars an hour
OpenRouter has it, you can test it for free using their chat interface, or for $5 if you want API access. No need to install or download anything.
Just one, and assuming no extra training? I think I’d go with Capybara Tess Yi 34b. In part because of how well it seems to follow instructions. But also because it has the broadest scope of knowledge that I’ve seen in any of the models so far. A lot of the models tap out on a lot of things past what you’d get from the first paragraph of Wikipedia. I get that feeling far less often with capy so far.
This!
OpenHermes-2.5-Mistral-7B-16k imo
Wizard-lm 13b
Mistral has been quite good at multiple tasks I throw at it given its small size. But for specific tasks some models can work better
I’m still shocked at how good mistral is. I wrote it off as a meme model for far too long just because of how overstated the praise seemed to be. But the thing really is amazing for the size.