davidbarker

A possible solution may be to incorporate artificial images in the training data. So, create an initial LoRA with the original Degas images and generate 500 images. From those generated images, pick the ones that most resemble Degas. Add those to the training set and train again. Repeat until (hopefully) it learns the correct style.

If you do end up training one on yourself with fal, it should ultimately take you here (https://fal.ai/models/fal-ai/flux-lora) with your new LoRA pre-filled.

Then:

1. Click 'Add item' to add another LoRA and enter the URL of a style LoRA's SafeTensor file (with Civitai, go to any style you like and copy the URL from the download button) (you can also find LoRAs on Hugging Face)

2. Paste that SafeTensor URL as the second LoRA, remembering to include the trigger word for yourself (you set this when you start the training) and the trigger word for the style (it tells you on the Civitai page)

3. Play with the strength for the LoRAs if you want it to look more like you or more like the style, etc.

-----

If you want a style LoRA to try, this one of SNL title cards I trained actually makes some great photographic images. https://civitai.com/models/773477/flux-lora-snl-portrait (the download link would be https://civitai.com/api/download/models/865105?type=Model&fo...)

-----

There's a lot of trial and error to get the best combinations. Have fun!

I have a MacBook Air so I train using the various API providers.

For training a style, I use Replicate: https://replicate.com/ostris/flux-dev-lora-trainer/train

For training a concept/person, I use fal: https://fal.ai/models/fal-ai/flux-lora-fast-training

With fal, you can train a concept in around 2 minutes and only pay $2. Incredibly cheap. (You could also use it for training a style if you wanted to. I just found I seem to get slightly better results using Replicate's trainer for a style.)

A reasonable amount of training images (50 or so), and then I train for 2,000-ish steps for a new style.

Many of them work well with Flux, particularly if they're illustration-based. Some don't seem to work at all, so I didn't upload those!

Here are a few in Degar style I made after training for 2,500 steps. I'd love to hear what you think of them. To my (untrained) eye, they seem a little too defined, perhaps?

https://imgur.com/a/sqsQLPg

Here are a few I've recently trained: https://civitai.com/user/dvyio

https://github.com/upstash/semantic-cache

  "Semantic Cache is a tool for caching natural text based on semantic similarity. It's ideal for any task that involves querying or retrieving information based on meaning, such as natural language classification or caching AI responses. Two pieces of text can be similar but not identical (e.g., "great places to check out in Spain" vs. "best places to visit in Spain"). Traditional caching doesn't recognize this semantic similarity and misses opportunities for reuse."

Unfortunately I can't change the original link. Perhaps dang can.

https://www.youtube.com/@tesla

1. On-device AI

2. AI using Apple's servers

3. AI using ChatGPT/OpenAI's services (and others in the future)

Number 1 will pass to number 2 if it thinks it requires the extra processing power, but number 3 will only be invoked with explicit user permission.

[Edit: As pointed out below, other providers will be coming eventually.]

You could use something like Little Snitch (on Mac) to check if it makes any calls to their servers.

They also allow you to override the URL for the OpenAI models, so although I haven't tried, perhaps you can use local models on your own machine.

It's a fork of VS Code with some AI features sprinkled in. It writes around 80% of my code, these days.

It also has a few useful features:

- a chat interface where you can @-mention files, folders, and even documentation

- if you edit a line of code, it suggests edits around that line that are useful (e.g. you change a variable name and it will suggest updating the other uses, which you accept just by pressing Tab)

- as you're writing/editing code, it will suggest where your cursor might go next — press Tab and your cursor jumps there

https://x.com/openaidevs/status/1779922566091522492

https://replicate.com/philz1337x/clarity-upscaler https://replicate.com/philz1337x/multidiffusion-upscaler

However, this weekend someone released an open-source version which has a similar output. (https://replicate.com/philipp1337x/clarity-upscaler)

I'd recommend trying it. It takes a few tries to get the correct input parameters, and I've noticed anything approaching 4× scale tends to add unwanted hallucinations.

For example, I had a picture of a bear I made with Midjourney. At a scale of 2×, it looked great. At a scale of 4×, it adds bear faces into the fur. It also tends to turn human faces into completely different people if they start too small.

When it works, though, it really works. The detail it adds can be incredibly realistic.

Example bear images:

1. The original from Midjourney: https://i.imgur.com/HNlofCw.jpeg

2. Upscaled 2×: https://i.imgur.com/wvcG6j3.jpeg

3. Upscaled 4×: https://i.imgur.com/Et9Gfgj.jpeg

----------

The same person also released a lower-level version with more parameters to tinker with. (https://replicate.com/philipp1337x/multidiffusion-upscaler)

Disclosure: Professor Malan is a friend of mine, but I was a fan of CS50 long before that!

In particular, looking at the video titled "Borneo wildlife on the Kinabatangan River" (number 7 in the third group), the accurate parallax of the tree stood out to me. I'm so curious to learn how this is working.

[Direct link to the video: https://player.vimeo.com/video/913130937?h=469b1c8a45]

It uses GPT (I believe they're running their own fine-tuned version), and allows you to '@' items like your files, folders, and (most impressively) any documentation.

So, if I'm working on something that uses React Hook Form, I can paste the URL to the RHF documentation. Cursor will index it (presumably creating embeddings for each page), and then I just '@' the RHF documentation in the chat and it'll find the relevant pages and include it them the context for GPT to answer.

It's very useful. I already subscribe to ChatGPT Plus but Cursor adds something special that I'm more than happy to subscribe to Cursor too.

Then, over time, it started to feel like the people higher up were less and less interested in the site, and no amount of moderation could fix some of the issues like the amount of spam.

I had a brief moment a couple of years ago where I went back to the site, cleaned up the spam and low quality posts I could find in the top few pages in the hope that it might kick off some kind of resurrection. It didn't, mainly because hardly anyone except spammers were posting to the site at that point.

https://platform.openai.com/docs/guides/text-to-speech

The first that comes to mind: https://chrome.google.com/webstore/detail/superpower-chatgpt...

The docs say:

> By default, images are generated at standard quality, but when using DALL·E 3 you can set quality: "hd" for enhanced detail. Square, standard quality images are the fastest to generate.

https://platform.openai.com/docs/guides/images/usage

(Timestamp 33:26)

Edit: updated the timestamp

If you ask it for 4 images with an exact prompt, it'll generate 4 identical (or almost identical) images. Then if you ask it to re-run it with a different seed, it will say that it's done it but it'll still generate the same images.

I hope that's going to be changed in the future.