Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Trae: An AI-powered IDE by ByteDance

trae.ai

Post reply on HN
Print view View original

41–50 of 428 posts

Re: Trae: An AI-powered IDE by ByteDance

#41

Post by ed »

DoctorOW wrote: post #11

I feel like this would've been better as a VSCode extension. Copilot, Q, Gemini, all were able to take this approach. Also, VSCode isn't considered a full IDE and adding some AI features isn't enough to change that. It seems like they forked VS Code just for the ability to say they "created an IDE" in the same way other projects fork Chromium to "build a browser".

Nah the extension API is pretty limited. Copilot uses proprietary API’s not available to extensions.

If you really want an integrated experience, and not just a sidebar UI, you need to go the same route as Cursor and fork Code-OSS (the MIT-licensed part of VS Code, analogous to Chromium for Chrome)

Re: Trae: An AI-powered IDE by ByteDance

#42

Post by segmondy »

echelon wrote: post #7

China is taking every venture-funded AI startup and shitting on them. Almost all of them. Tencent's Hunyuan 3D - Tripo, etc. Tencent's Hunyuan Video - Sora, Runway, Pika, etc. ByteDance's Trae - Cursor, etc. DeepSeek R1 - GPT, etc. Unitree - Figure, Tesla, Boston Dynamics, etc. And a ton of this stuff is completely open source. All of the Tencent stuff is. This destroys the moat of so many companies that have spent h…

Trae doesn't look opensource, from their site looks like you download the binary (osx only for now) and install it. You can't configure it to use your own AI or another provider. You need to use their own API, all your prompts and code are go to bytedance.

Re: Trae: An AI-powered IDE by ByteDance

#43

Post by segmondy »

handfuloflight wrote: post #39
mring33621 wrote: post #8

What LLM provider(s) is it talking to?

According to the Discord, Claude 3.5 Sonnet and OpenAI GPT-4o. Also according to the Discord, "SPRING (SG) PTE. LTD. or one of its affiliates" is fully subsidizing the costs. This doesn't look like a bring your own keys type of deal.

If it's truly talking to OpenAI or Claude, then the traffic can be intercepted and redirected to your own AI router for other cloud AI providers or your local Ollama instance.

Re: Trae: An AI-powered IDE by ByteDance

#44

Post by nosefurhairdo »

toprerules wrote: post #26

I feel like I'm taking crazy pills, because every time I try to use AI for any of my projects - including the "better" Anthropic and OpenAI models, I get terribly written, buggy code that has nothing to do with the domain or question. Unless I'm feeding it actual softball, single function questions it adds almost no value to my dev cycle. Mind you, all of this terrible code also costs a ton in tokens. It's sad becaus…

It takes some practice to get good using them. I use it as a sophisticated autocomplete. For example, if I'm building an api client layer in a web frontend, I might have `fooApi.ts`, `mockFooApi.ts`, and `barApi.ts`. instead of writing `mockBarApi.ts`, something that might take 5 - 15 minutes, I can feed those 3 existing files into context, and a few seconds and pennies later I have a nearly perfect file ready for review.

For any semi novel development I want to write it myself to establish the patterns I like and get an understanding of what I'm building. It's the boring stuff I hand over to Claude.

Re: Trae: An AI-powered IDE by ByteDance

#45

Post by anon-3988 »

toprerules wrote: post #26

I feel like I'm taking crazy pills, because every time I try to use AI for any of my projects - including the "better" Anthropic and OpenAI models, I get terribly written, buggy code that has nothing to do with the domain or question. Unless I'm feeding it actual softball, single function questions it adds almost no value to my dev cycle. Mind you, all of this terrible code also costs a ton in tokens. It's sad becaus…

You don't use LLMs to write code for you. This is like asking it what is 2+2, it may be able to answer that but that's not how it works.

What it is just a summary of the internet.

Re: Trae: An AI-powered IDE by ByteDance

#47

Post by spondyl »

I know it's a marketing site so the speed is for the sake of the demo but the various animations showing a cursor scrolling past reams of text quicker than any human can scan, and just blindly hitting Accept makes for weird optics.

Re: Trae: An AI-powered IDE by ByteDance

#48

Post by opwieurposiu »

impulser_ wrote: post #40

Bro the West is getting cooked by China right now. We might be fucked. Everything that has been coming out of China is better than anything our mega trillion dollar corporation are producing and they are doing with with sanctions on the best GPUs. I don't work in big tech, why are they so bad with sigifically more money and more advanced technology? Google had a decade and a half head start and they can't even produc…

My kids are in elementary school that Chinese language in the morning and English in the afternoon.

Only problem is now I have no idea what they are talking about half the time.

Re: Trae: An AI-powered IDE by ByteDance

#49

Post by tnt128 »

impulser_ wrote: post #40

Bro the West is getting cooked by China right now. We might be fucked. Everything that has been coming out of China is better than anything our mega trillion dollar corporation are producing and they are doing with with sanctions on the best GPUs. I don't work in big tech, why are they so bad with sigifically more money and more advanced technology? Google had a decade and a half head start and they can't even produc…

Haven’t tried Trae. Is it objectively better than cursor? I felt this statement might be premature. US is still leading in software and AI space.

Re: Trae: An AI-powered IDE by ByteDance

#50

Post by segmondy »

softirq wrote: post #30

Side note, but I hate that we're moving to a world where coding costs a subscription. I fell in love with coding because I could take my dad's old Thinkpad, install Linux for free - fire up Emacs and start hacking without an internet connection. We're truly building walls everywhere.

No one is forcing you to subscribe. You can code the old fashion way, if you wish to use AI, you can run your own local model.
Post reply on HN