neal_

Neal here- Still very much under development. The public repo is now several months behind. I have been contracted to integrate my project into a particular product, which I have been working this last month. To do this I have been working on the embeddable version, which now can run in Unreal, Vulkan, and OpenGL. It also works on imgui docking branch. Anyways, hopefully I will be able to upstream all my work and make it public. I do intend to continue this project as time permits.

Hey Ned developer here, there is a target fps slider in settings. I am able to get 800+fps on my MacBook m4. There is a mode where it reduces fps until user input on Mac to save battery life. However this only works with shaders turned off, as the shaders depend on continuous rendering.

:p

The better the benchmarks, the worse the model is. Subjectively for me the more advanced models dont follow instructions, and are less capable of implementing features or building stuff. I could not tell a difference in blind testing SOTA models gemini, claude, openai, deepseek. There has been no major improvements in the LLM space since the original models gained popularity. Each release claims to be much better the last, and every time i have been disappointed and think this is worse.

First it was the models stopped putting in effort and felt lazy, tell it to do something and it will tell you to do it your self. Now its the opposite and the models go ham changing everything they see, instead of changing one line, SOTA models rather rewrite the whole project and still not fix the issue.

Two years back I totally thought these models are amazing. I always would test out the newest models and would get hyped up about it. Every problem i had i thought if i just prompt it differently I can get it to solve this. Often times i have spent hours prompting starting new chats, adding more context. Now i realize its kinda useless and its better to just accept the models where they are, rather then try and make them a one stop shop, or try to stretch capabilities.

I think this release I won’t even test it out, im not interested anymore. I’ll probably just continue using deepseek free, and gemini free. I canceled my openai subscription like 6 months ago, and canceled claude after 3.7 disappointment.

hahahahaha

thats crazy! I haven't heard of yolo mode?? dont they like restrict access to the project? but i guess the terminal is unrestricted? lol i wonder what it was trying to do

Im building retro style text editor with Open GL shaders, its a work in progress. I have ported st.c to imgui and have mostly working terminal emulator, now im working on LSP adapters and extending language support. https://github.com/nealmick/ned

I was using gemini 2.5 pro yesterday and it does seem decent. I still think claude 3.5 is better at following instruction then the new 3.7 model which just goes ham messing stuff up. Really disappointed by Cursor and the Claude CLI tool, for me they create more problems then fix. I cant figure out how to use them on any of my projects with out them ruining the project and creating terrible tech debt. I really like the way gemini shows how much context window is left, i think every company should have this. To be honest i think there has been no major improvement beyond the original models which gained popularity first. Its just marginal improvements 10% better or something, and the free models like deepseek are actually better imo then anything openai has. I dont think the market can withstand the valuations of the big ai companies. They have no advantage, there models suck worse then free open source ones, and they charge money??? Where is the benefit to there product?? People originally said the models are the moat and methods are top secret, but turns out its pretty easy to reproduce these models, and its the application layer built on top of the models that is much more specific and has the real moat. People said the models would engulf these applications built ontop and just integrate natively.

Cursor has no models, they dont even have an editor its just vscode

Working on a chrome extension that can message live support on a bunch of websites. Recently got a new version through app review, and have a pricing model now. Let me know what you think!

Working on a chrome extension that can message live support on a bunch of websites. Recently got a new version through app review, and have a pricing model now. Let me know what you think!

Appreciate it! Yea that video is pretty bad from a marketing perspective. I was kinda just thinking technical proof of concept shows it reading and replying...

Buttons and other actions are a bit tough. Currently it extracts the inner html from the chat wrapper div and gets all the text from the chat. This is great for being universal for different providers but it does have trade offs. detecting new messages can be tough if the inner html changes alot without an actual reply being received.... And depending on how the messages are labeled it sometimes has no way of detecting who is who, it relies mainly just off context.... Lots to improve. Overall supporting providers is good approach get alot of different website compatibility for integrating a single provider. So long as the customizations dont break anything. Main thing is making everything universal for it work on any site. That's the main challenge supporting lots of sites while still providing good context and the information needed. Appreciate the advice! I may post again when i have some of this stuff dialed in

I just put this together to save time and effort. It's not a product because its free currently. The video is just showing how it works, I will make better videos and full length! it usually cant do it all, but with a little help its perfect for certain situations.

order delivery on 2 phones and let them talk to each other

Working on a chrome extension that connects GPT to a bunch of companies support, like xfinity and geico.