Thank you ! PoW is the only way!

We are constantly monitoring for the latest and greatest models. If a new model performs better, we deploy it! We are also tracking new things (Vision is a good example) and if we see value in it, we put it on the platform.

But to answer your question specifically:

Text:

• Mixtral 8x7B is currently the best Open-Source model based on our internal tests as well as multiple benchmarks. It's also very efficient which is what always us to charge only 21 sats per prompt.
• We also offer a "Code" Model called Call-Llama2 70b, which can produce better results than GPT-4 on this specific task.
• We are looking into adding another totally uncensored model, where no subject/topics are off-limits.

Image:

• Stable Diffusion is for now the best Open-Source image model. Other models exist that could be cheaper, but given the performance and limits (see above comment) of the best model, we don't think they actually bring a lot of value.

Vision

• LLaVA is an incredible model that came out just a few days after GPT-Vision and is the best we tested so far. Multimodal is key to unlocking new use cases to LLM.

Audio:

• This is more of a "toy" model. It's fun to try but is still very limited in its current form. We just wanted to put it out there so people can see another "side" of AI.

Thanks! Agreed that expectations need to be tempered around text to image. You need to be extremely descriptive about what you want and likely will need several attempts, but eventually you will get a brand new image free of copyrights that you can use at will in presentations, websites, social-media, etc...

Open-Source models are key and if you have the appropriate hardware, know-how and time to run it yourself, this is the best approach. Most people, however, don't at this point and this is what we are looking to address.

Agreed with you about "intelligence". I still see value in what it can do at this point, copying here my response to a similar comment above:

"I still see a lot of value in the current tools, most notably when looking to merge multiple well-documented concepts and sources. Ex: Who was the mayor of Paris during WW2? Create an itinerary for 2 young people budget-conscious in Rome for 2 days, travelling with a Vespa and staying at hotel "X". Create a list of all the "rare" languages spoken in Africa, including for each an approximation of how many people speak that language and in which country "Search" would not give you a direct answer to any of those questions. It would require looking at multiple websites and taking notes. Obtaining such an answer almost instantly for 21 sats is, to me, incredible value at this point. I think we just need to keep playing with it until we find which use cases is it good for."

I understand. The name "AI" creates certain expectations that are not always met at this point. I still see a lot of value in the current tools, most notably when looking to merge multiple well-documented concepts and sources. Ex:

1. Who was the mayor of Paris during WW2?
2. Create an itinerary for 2 young people budget-conscious in Rome for 2 days, travelling with a Vespa and staying at hotel "X".
3. Create a list of all the "rare" languages spoken in Africa, including for each an approximation of how many people speak that language and in which country

"Search" would not give you a direct answer to any of those questions. It would require looking at multiple websites and taking notes. Obtaining such an answer almost instantly for 21 sats is, to me, incredible value at this point. I think we just need to keep playing with it until we find which use cases is it good for.