geepytee

The era of drawing machines already came and went, and it wasn't 3D printers that killed them.

I like it! Would like to have some sort of map/visualization of what topics i've explored for when diving into parallel sessions. Right now the UI only supports diving into one branch at the time.

Lots of them. Check out https://double.bot for example.

Hey HN! I'm the co-founder of Double (YC W23). Earlier this year we decided to pivot our company and walk away from $5k, and we wanted to share our journey, the rationale behind our decisions, and the lessons we learnt along the way.

Also happy to answer any questions about it! :)

Just added Llama 3 70B to Chat at https://double.bot if anyone wants to try it for coding. Free for your first 50 messages and will push Llama 3 8B for autocomplete later this afternoon.

From my early tests this seems like the first API alternative to GPT4 / Opus. Huge!

Also added Llama 3 70B to our coding copilot https://www.double.bot if anyone wants to try it for coding within their IDE and not just chat in the console

I was particularly excited for the high HumanEval score, and this is before the 400B model and the CodeLlama tune!

I just added Llama 3 70B to our coding copilot https://www.double.bot if anyone wants to try it for coding within their IDE

One of the OpenAI PM's was also saying the model got substantially better at math: https://x.com/owencm/status/1777770827985150022

I'm trying it for coding and have added it to my VS Code Copilot extension Overall I'd say it's better at coding than the previous GPT-4 Turbo. https://double.bot if anyone wants to try it :)

We implemented the subscription a few hours ago, and still hold a very generous free tier. Unfortunately, we learnt that it's not possible to have an uncapped free tier, we saw several abusers use our API to run their own apps which is obviously against the spirit of a coding copilot :)

JetBrains plugin is on the roadmap, soon!

Really cool to think about it. I didn't think they'd just drop 2 tonnes worth of batteries like that.

Hey, co-founder of https://double.bot coding copilot here. Thanks for sharing this! I'm assuming OP is the author?

Curious how does the double quantization of weights in qLORA impact the model's performance compared to standard fine-tuning. Could qLORA and LoRA be combined with other optimization methods, such as pruning or knowledge distillation, for more efficiency gains?

Hey! Founder of https://double.bot here (unfortunately we don't have a neovim extension yet), we're also a Canadian team. Thanks a ton for sharing this!

Out of curiosity, what got you to finally try a Copilot after all of these years? Also where can I learn more about using AI to sell items on Kijiji? I really liked your point on asking copilot to suggest code improvements, I wonder if we could make this proactively.

This is great, going to incorporate this to our roadmap.

What else should an IDE integrated coding copilot be able to do?

How do I get paid to use LLMs? That's the dream.

Wasn't aware that it's not widely available but if you can't find it anywhere else, we have it available in our free tier on Double: https://double.bot

So basically your db says Claude 3 Opus is the best coding LLM, correct?

Have you ran into issues with Copilot where it does things like:

- Doesn't close brackets / adds too many closing brackets

- Spams completions while you're writing comments

- Doesn't auto import the required libraries after accepting a suggestion

- No multi cursor mode

- Refuses to name variables

- Refuses to trigger in the middle of a line

- Lower quality code than ChatGPT (probably because they use a smaller model)

Asking because I was also a heavy Github Copilot user but after 1-2 years of constant subpar UX and seemingly no effort from the Copilot team to fix it, I just went ahead and built my own extension that fixes all of the above: https://double.bot

Also it wasn't on your list but you mentioned GPT-4 is smart. You should try the new Claude 3 Opus, we implemented it on Monday and have been getting super positive feedback from users!

When you say noise do you mean the autocomplete just spamming you all the time, like for example in the middle of writing a comment?

Software dev is one of those careers where top performers earn a lot of money, so IMO a tool that generates higher quality code is a no brainer, even if it's 'expensive'. Similarly to a Bloomberg terminal if you are in IB (those things are $25k/mo).

That being said if you want to keep playing with Claude 3 Opus, we have it available on our free tier at https://double.bot

The issue with local IMO is that you'll going to have to compromise on quality. And I really don't want to waste my time talking with inferior models when better models are widely available.

Agreed that Github Copilot is not worth anyone's valuable time. You should check out the new Claude 3 Opus model, it's noticeably better. Right off the bat, it's less 'lazy' with its generations, and for me, it has been able to solve bugs that GPT-4 could not solve.

Just this week we made it available on https://double.bot (VS COde Copilot extension). Have been getting similar feedback from multiple users

I was asking myself the same question not so long ago because Github Copilot's bugs kept bothering me and this is what got us to go into YC and build https://double.bot

If what you are looking for a high quality UX + access to the most capable models (today that'd be the GPT-4 killer, Claude 3 Opus), then I think you'll like what we've built.

Double ha similar features to all the other Copilots, code autocomplete, chat, etc but we've put particularly care around getting the small details right (even the smallest thing like making sure our autocomplete closes the brackets appropriately). If you're coming from Github Copilot you should check out this side-by-side comparison: https://docs.double.bot/copilot

+1, Claude 3 Opus is a game changer.

Since OP asked about a Copilot product, we (YC W23) actually built https://double.bot to do exactly this: high quality UX + most capable models. Although we don't have a subscription right now, it's free for the time being :)

I've been coding with Claude 3 Opus since it first launched this past Monday. From the start I thought it was better than GPT-4, on a surface level it was less 'lazy' with its completions, and as I dived deeper, it was able to produce good code to solve bugs GPT-4 Turbo was not able to do.

IMO Opus is hands down the superior model for coding, the GPT-4 killer if you will. We immediately shipped it to all of our users at https://www.double.bot (coding copilot) and have been hearing similar stories from users all week.

I used to be the guy who would save cool websites I stumble across so I could 'read them later'. Lol, ya right.

Fast forward to today, I've been using Zenfetch for a few months now. Basically, everything I read, or wish I had the time to read / watch (they support Youtube too) gets saved on my Zenfetch library.

Being able to ask for summaries of stuff you are browsing in real time is great for picking up new knowledge.

Chatting with your entire library with 1,500+ saved articles and drawing correlations between different subjects + never forgetting anything, is next level.

Don't be the 'bookmarking this for later' guy.

I watch a bunch of MIT OCW classes on youtube. Is there a limit on how long the Youtube videos can be in order to save?

As a 20 something y/o, jealous of the 50 y/o's going thru a midlife crisis who can afford this!

AFAIK this is a first, GPT not only writes the lyrics (which ChatGPT has done for months now), but also generates MP3 files of the finished songs.

The BYD team has done an insane job at ramping up, and is likely to overtake Tesla on Q4 2023 according to the latest production and sales mumbers.

Don't we all have a similar anecdote to tell?