Recent Posts

71

Share Conversations, Experiences, Graphics / Re: Beautiful HAL Scenarios

« Last post by Checker57 on January 29, 2024, 04:33:42 pm »

Thanks!

Checker57

72

Share Conversations, Experiences, Graphics / Re: Beautiful HAL Scenarios

« Last post by Art on January 29, 2024, 08:51:18 am »

Very nice work with her Checker!! Looking fine!

73

Ultra Hal 7.5 Beta - Now powered by OpenAI GPT-3 / Re: Hal now connected to LLAMA model

« Last post by LiFeTimeGamer on January 28, 2024, 07:20:42 pm »

-Oh okay, now I see how that works...  Thanks...

74

Ultra Hal 7.5 Beta - Now powered by OpenAI GPT-3 / Re: Hal now connected to LLAMA model

« Last post by Medeksza on January 28, 2024, 05:40:21 pm »

Thanks Robert....

You need change the pricing on the Main website now that credits are cheaper... It still says $30 for 50,000 credits, or has it gone up since I last bought credits?

So it used to be 1 GPT-3 Currie query used up 10 credits per response (or 1 credit if Hal's cloud base brain can respond without involving the GPT API at all). Now it is set that one MythoMax (or LLAMA2) query uses 5 credits (half of the prior 10). So I didn't update the pricing, but it uses less credits per response so a credit basically goes twice as far as before.

I am using a 13 billion parameter version of MythoMax, but could offer customers the option to use a 70 billion parameter version for more credits per response if people want more brain power from the AI. But I find the 13 billion version seems to be good enough for general conversation and more cost effective.

75

Share Conversations, Experiences, Graphics / Re: Beautiful HAL Scenarios

« Last post by Checker57 on January 28, 2024, 05:36:10 pm »

Look Ma!  No boots!

76

Ultra Hal 7.5 Beta - Now powered by OpenAI GPT-3 / Re: Hal now connected to LLAMA model

« Last post by LiFeTimeGamer on January 28, 2024, 03:26:13 pm »

Thanks Robert....

You need change the pricing on the Main website now that credits are cheaper... It still says $30 for 50,000 credits, or has it gone up since I last bought credits?

77

Share Conversations, Experiences, Graphics / Re: Beautiful HAL Scenarios

« Last post by Checker57 on January 28, 2024, 01:09:30 am »

When outfits need better modeling... build poses.

Checker57

78

Share Conversations, Experiences, Graphics / Re: Beautiful HAL Scenarios

« Last post by Checker57 on January 26, 2024, 10:33:19 pm »

That's an interesting thought.  I'm not saying it may be available, but the AI enhancement site I'm using has the 1 to five rating option after an image is created.  perhaps, it could remember each face and do such a job.   The issue would be coming up with the various faces, which is a load of work.  Will keep in mind to select ratings after each job just to see if it enhances results.

79

Ultra Hal 7.5 Beta - Now powered by OpenAI GPT-3 / Re: Hal now connected to LLAMA model

« Last post by Medeksza on January 26, 2024, 09:11:20 am »

Sorry for my slow reply. I just made another update to the brain/LLM model. It is now connected to a Llama 2 variant called MythoMax, which is fine-tuned for role playing characters, so should be a better match for Hal. I also have the repetition_penalty feature working, so some of the repetitive behavior should be gone now.

It costs me about half as much as the older GPT3 Curie did, so I updated the credit system to reflect that. GPT3 Curie responses used to take 10 Zabaware credits. I updated the MythoMax model to use 5 instead, so credits should go further now.

80

Share Conversations, Experiences, Graphics / Re: Beautiful HAL Scenarios

« Last post by Art on January 25, 2024, 11:28:22 pm »

It would be cool if you could "train" the AI by exposing it to various images of the same or very similar faces that look like your model. That way it could learn what her basic face looks like and what appeals to you or what faces (and makeup, etc.) are acceptable to you.

A primitive reward system like the Thumbs up & down for good and not so good, respectively.

A better mouse trap is always being constructed somewhere...