rgbrgb

14 ms·

211.
by rgbrgb 3y ago
0 chance I would have tried this without the playground. Now thinking about where I can use it. Nice tool!
212.
by rgbrgb 3y ago
> Unlike many traditional SSGs where each navigation results in a full page reload, a website generated by VitePress serves static HTML on the initial visit, but becomes a Single Page Application (SPA) for subsequent navigation within th
213.
by rgbrgb 3y ago
I think it was similar... an agent finds a bunch of info on a private/public company to evaluate investment (automating some associate analyst work). TBH I'm not sure where they ended up but I know they had an interesting distribu
214.
by rgbrgb 3y ago
This is so cool, congrats on the launch! I have a friend who looked into doing something similar but they couldn't figure out a way to get the cost low enough. This was like a year ago so I'd guess it's much cheaper now and y
215.
by rgbrgb 3y ago
Congrats on launching! I don't quite understand what this solves relative to vite. One sentence on that in the README would probably go a long way.
216.
by rgbrgb 3y ago
Correlates with the relatively high number of executives I know who do not drink alcohol.
217.
by rgbrgb 3y ago
Thanks for releasing this! What is your use case for this rather than llama.cpp? For the on-device AI stuff I mostly do, llama.cpp is better because of GPU/metal offloading.
218.
by rgbrgb 3y ago
At first glance this looks very similar to just adding the contrastive prompts to the beginning of the system prompt to "prepare" the logits. What am I missing?
219.
by rgbrgb 3y ago
i don’t buy that premise. in practice we’re seeing a lot of evidence that you can’t trust the open evals because of contamination (maybe accidental, though there’s definitely incentive to cheat and move up the leaderboards). closed/sub
220.
by rgbrgb 3y ago
yea, that's my first thought seeing the result too. we need a reputable proprietary eval.
221.
by rgbrgb 3y ago
Supports commercial use! Interesting what's unsupported: - In any way that violates any applicable national or international law or regulation or infringes upon the lawful rights and interests of any third party; - For military use in
222.
by rgbrgb 3y ago
I know chat is lame and overdone but here's my open source local AI chat app for macOS :). I wanted something simple enough for the non-technical people in my life who were using ChatGPT. For better or worse, those people are mostly no
223.
by rgbrgb 3y ago
everyone (not just programmers) should learn SQL, and I have many times! For me, programming languages are pretty different from bikes in that I have very little "muscle memory" for syntax when staring at a blank page, especially
224.
by rgbrgb 3y ago
with data privacy decisions I like to think about how the policy/reasoning would sound in a NYT headline or congress testimony. your stance does not imbue trust for me. 0 employment contracts or user facing ToS I've read make a di
225.
by rgbrgb 3y ago
about as much as I trust 23andme to keep my genetic data private i give their intent the benefit of the doubt but I’ve been on the other side of too many data collection systems to trust that theirs is foolproof and more secure than my loca
226.
by rgbrgb 3y ago
So it looks like it scores 76.5% on SQL-Eval [0], a bit behind GPT-4 at 83% and sqlcoder-15b at 78%. What kind of applications would this be useful for? What can you build with an AI data science intern that's right 75% of the time? A
227.
Make Word Art(makewordart.com)
1 points by rgbrgb 3y ago|0 comments
228.
by rgbrgb 3y ago
SEEKING VOLUNTEERS: https://github.com/psugihara/FreeChat FreeChat is a simple LLM chat app built with SwiftUI and llama.cpp server. The main goal of FreeChat is to make open, local, private models accessible to more p
229.
by rgbrgb 3y ago
I will speculate! They have a model that far surpasses GPT-4 (achieved agi internally) but sama is back on a handshake agreement that they will only reveal models that are slightly ahead of openly available LLMs. Their reasoning being that
230.
by rgbrgb 3y ago
I love it and agree with other commenters on animations. Any traction on subscriptions? I'm always curious if there's a biz model that works for this kind of game on web (besides ads). It's funny because I think you could abs
231.
by rgbrgb 3y ago
love this! The sounds are great and the caching is smart. I'd add a little loading animation or something... as a dev I knew it was hitting the LLM but my friend thought it was just laggy. I prototyped a similar concept with a numi&#x2
232.
WhisperKit: Speech Recognition On-Device for iOS and macOS(github.com)
2 points by rgbrgb 3y ago|0 comments
233.
by rgbrgb 3y ago
big improvement! good luck
234.
by rgbrgb 3y ago
Congrats on the launch. Looks interesting. Quick thoughts on the landing page: - Pricing looks awesome. - I'm not currently the target audience because everything I'm doing right now is open source with free GitHub actions. - I&#x
235.
by rgbrgb 3y ago
Quantization can take it under 30GB (with quality degradation). For example, take a look at the GGUF file sizes here: https://huggingface.co/TheBloke/Llama-2-70B-GGUF
236.
by rgbrgb 3y ago
This is a little lib I love and was surprised to see it had just 15 stars. Great, simple async API. It's an important dependency for my LLM chat app, FreeChat, and the LLM-based game I'm working on now.
237.
EventSource: Server-Sent Events in Swift(github.com)
2 points by rgbrgb 3y ago|1 comments
238.
by rgbrgb 3y ago
Are these libraries for connecting to an ollama service that the user has already installed or do they work without the user installing anything? Sorry for not checking the code but maybe someone has the same question here. I looked at usin
239.
by rgbrgb 3y ago
I've been looking at this one to embed in desktop apps https://github.com/unum-cloud/usearch
240.
by rgbrgb 3y ago
I quit amazon as soon as I had enough to live on (very frugally) for 18 months and a team who was willing to go full-time on our app idea. We had made a little prototype but not released and no traction or biz model. A lot of my confidence

More ›