TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

Rabbit R1 source code [part 1]

353 pointsby wibbilyabout 1 year ago

34 comments

ado__devabout 1 year ago
Not surprised at all. This and the Humane Pin both seemed like a quick cash grab before phones integrated all the new AI goodness. I'm expecting we'll see that window close by the time I/O and WWDC wrap up this year, so they had to move fast.
评论 #40136552 未加载
评论 #40139293 未加载
评论 #40135833 未加载
评论 #40135799 未加载
ipsum2about 1 year ago
I implemented my own DIY version of Rabbit at a hackathon using Playwright and VNC. I feel extremely validated that they use the same things that I thought of.
评论 #40140487 未加载
NetOpWibbyabout 1 year ago
According to their CTO in the Discord[0]:<p>&gt; If someone spends enough time with the login minions they can extract these code. But these code are locked down and are sanitized. LAM lives elsewhere. This is someone looking at the rabbit hole not understanding how it works. And tries to be smart.<p>[0]: <a href="https:&#x2F;&#x2F;cdn.discordapp.com&#x2F;attachments&#x2F;1185274946981732374&#x2F;1232410511652687902&#x2F;IMG_8101.png?ex=66295b3b&amp;is=662809bb&amp;hm=929befc438ea4ca2e7d459a1ef1b2a3caa562e9d0597e8472316af396672e021&amp;" rel="nofollow">https:&#x2F;&#x2F;cdn.discordapp.com&#x2F;attachments&#x2F;1185274946981732374&#x2F;1...</a>
评论 #40138037 未加载
extrabout 1 year ago
I really think this + the humane AI pin would be super interesting products if they made them hackable. The hardware is super cool, no problem if the software isn&#x27;t there yet, it&#x27;s not like they&#x27;re being sold at Best Buy, I&#x27;m not worried about my mom acquiring one of these and getting her passwords leaked. Who cares if the auth flow is super hacky&#x2F;insecure? Let us self host it! Let the community create more playwright scripts!
评论 #40135737 未加载
评论 #40135959 未加载
评论 #40137452 未加载
评论 #40135765 未加载
评论 #40139858 未加载
rodigerabout 1 year ago
At least it&#x27;ll look nice on a shelf thanks to Teenage Engineering&#x27;s good work :)
评论 #40136049 未加载
评论 #40135655 未加载
评论 #40135901 未加载
amiantosabout 1 year ago
I&#x27;m really into AI stuff but both the AI Pin and Rabbit R1 underwhelm. They are products that don&#x27;t need to exist if the problem they&#x27;re solving was truly solvable right now, because the best place for that problem to be solved is already in our hands: our phones. But we&#x27;re not all talking to AI assistants in our phones all day. Why? Because the technology isn&#x27;t good enough to do it yet? Because people don&#x27;t want to talk to digital assistants? Once the tech is good enough that it can motivate ordinary people to look silly talking to their phones outloud, it&#x27;ll be on our phones and easy to use, and there will be no need for kitschy little handheld devices. No one wants to carry around another device.<p>That said, smart glasses sound like a great idea to me, but I wear glasses all ay long, so I am extremely biased. I don&#x27;t think most people want to voluntarily wear glasses to just put a computer on their face, so I wouldn&#x27;t bet on glasses, either. Sorry, Zuck.
评论 #40136500 未加载
评论 #40137505 未加载
评论 #40141147 未加载
评论 #40136410 未加载
phhabout 1 year ago
&gt; In reality, they&#x27;re simply relying on several Playwright automation scripts to do the job for you, which is why they only support four apps: Spotify, Midjourney, Doordash, and UberEats.<p>I think that part is mostly fine? I&#x27;d rather make give a LLM access to <a href="https:&#x2F;&#x2F;woob.tech" rel="nofollow">https:&#x2F;&#x2F;woob.tech</a> to be my personal assistant while parsing 99% less noise, than have a LLM that parse and understand stupidly complicated web pages, and randomly fail at the task because the name of my doctor is bobby drop tables.<p>That being said, it can be interesting to use LLMs to assist creating woob plugins.
评论 #40136859 未加载
评论 #40136572 未加载
jthnmeabout 1 year ago
Some will call it scam, some will call it MVP
评论 #40135847 未加载
评论 #40137168 未加载
评论 #40135878 未加载
xoriabout 1 year ago
Not much here explicitly in the source code dump. A little insight into their worker node infra but no &quot;secret sauce&quot; imo.
评论 #40136846 未加载
animexabout 1 year ago
I just couldn&#x27;t fathom the big three phone platforms not implementing this on a device that we all have and is capable of same if-not better dynamic voice integration.<p>At the very least, I hope products like the Rabbit spur these companies to start innovating again. Even if they are smoke &amp; mirrors, the interest shows there&#x27;s demand for these features.<p>Site Note: I&#x27;ve noticed Google Home&#x27;s voice assistant has declined over time -- it used to handle complex queries and now it can barely understand simple directions. It used to understand me perfectly in the noisiest environments and now it makes many transcribing errors.
评论 #40136310 未加载
Jonovonoabout 1 year ago
What NFT projects were they involved with?
评论 #40136012 未加载
abrichrabout 1 year ago
For an open source alternative, check out <a href="https:&#x2F;&#x2F;github.com&#x2F;OpenAdaptAI&#x2F;OpenAdapt">https:&#x2F;&#x2F;github.com&#x2F;OpenAdaptAI&#x2F;OpenAdapt</a>. We combine Segment Anything Model with GPT4-V to understand recordings of workflows in desktop user interfaces, then replay them according to the user&#x27;s instructions.
ach9labout 1 year ago
why are people surprised they use playwright? how do they imagine a lam would work? in the end an &quot;action&quot; would always have to come as a computer command, be it an api call, or function call generation, an &quot;action&quot; is code to be executed. transformers generate tokens and only tokens, it is up to you to decide how you want the flow of tokens to be. i find the use of playwright quite clever, you could use puppeteer and make it write scripts for web browsing as well, but in the end an action will always come back to computer code, written as text as a human would do. the fact they use playwright does not imply it is not generative ai, on the contrary it is a clever way of showcasing how to configure a transformer to bypass an api lock-in such as the midjourney one.
adlpzabout 1 year ago
Any background on how the code got leaked? Insider? Hacked servers?
stranded22about 1 year ago
They gave 12 months perplexity pro with it - and I am already a subscriber. So, I basically paid a bit extra for another 12 months and a rabbit r1 to play with.<p>If it doesn&#x27;t work how I want, I should be able to sell on whilst keeping the perplexity pro sub.
评论 #40146210 未加载
latentcallabout 1 year ago
I thought the idea of the Rabbit R1 was cool. I have a strong feeling this will end up like the Humane Pin, which is sad. I&#x27;m glad companies are trying something different.<p>I&#x27;d love to be able to use my phone hands free without having to look at it, and interface with ChatGPT&#x2F;Claude&#x2F;whatever but I am not sure if it&#x27;s possible? Siri works very poorly and is unreliable. I&#x27;d like to be able to use an LLM as a personal assistant. Set timers, call people, message people, but also be able to ask questions like the voice chat function in the ChatGPT app. Maybe one day!
评论 #40138114 未加载
serfabout 1 year ago
I never understood the appeal outside the cute form-factor, all of the demos were absolutely terrible.<p>an aside : npr doesn&#x27;t like the &#x27;spade&#x27; comment, although I think the explanation is kind of iffy.[0]<p>[0]: <a href="https:&#x2F;&#x2F;www.npr.org&#x2F;sections&#x2F;codeswitch&#x2F;2013&#x2F;09&#x2F;19&#x2F;224183763&#x2F;is-it-racist-to-call-a-spade-a-spade" rel="nofollow">https:&#x2F;&#x2F;www.npr.org&#x2F;sections&#x2F;codeswitch&#x2F;2013&#x2F;09&#x2F;19&#x2F;224183763...</a>
Aleksdevabout 1 year ago
Rabbit R1 was such a dumb product. You have to buy a proprietary piece of plastic to use some flavor of chat GPT.<p>Never encountered a person with one.
vunderbaabout 1 year ago
One of the bizarre talking points in defense of the existence of rabbit was to get away from our phones. It&#x27;s just completely inexplicable to me because it&#x27;s not like it was ever intended to be a replacement, the only difference is congratulations you now have to lug around two separate brick shaped appliances wherever you go...
eterpstraabout 1 year ago
Maybe they hacked together something that can feasibly me marketed as an AI-assistant knowing that whatever they build now will get &quot;steamrolled&quot; by GPT-5 (Sam&#x27;s words, not mine). When GPT-5 gets released, update the OS and it&#x27;ll work as advertised... EZ-PZ!
mafuyuabout 1 year ago
Pretty much in-line with my expectations. I ordered one because I thought the design was neat, and I was interested in hacking around and flashing my own stuff onto it. The pricing was clearly at or below cost.<p>Looking at just the concept (and ignoring execution), I don&#x27;t really see the point of this thing? The whole thing is a feature that could exist on a smartphone. The dream of an AI agent that you can converse with to replace your smartphone could be compelling, but nowhere close to reality yet. Even then, the big smartphone OS companies are obviously better positioned for this. The smartphone is the hub for all your information, plus they have years of voice assistant, automation, and home IoT integration to build off of.<p>Humane was silly because it was a smartwatch without any of the proper software support, but Rabbit is essentially doing the same but targeting a smartphone replacement. If you really want to break out and try to dethrone smartphone vendors, you&#x27;ll have to come up with something more compelling than a worse user interface to a poorly made software platform. That&#x27;s a software feature you&#x27;re building.<p>In some sense, I do think Rabbit had a better approach than Humane, though. Getting a bunch of low-priced &quot;toy&quot; devices into the market that are just a frontend to your server software could get you off the ground. The software needs to exist, though...
iamleppertabout 1 year ago
Who cares how its made if you can make a bag from it before anyone is the wiser? The point in all this stuff is to make bags of money, whatever way you can do that don&#x27;t matter as long as you gettin the bread.
popeyabout 1 year ago
Looks like the repo was just removed from GH.
评论 #40147930 未加载
GaggiXabout 1 year ago
We already know that the large action model would not be available at launch, but I wonder how well it does works as an AI assistant.
m3kw9about 1 year ago
The UIUX isn’t good, AI models are useless without good user experiences
aw4yabout 1 year ago
I expect it (or at least I hope) to be really hackable.
评论 #40137588 未加载
ugh123about 1 year ago
If it can do what it claims to do, which is automate on top of existing apps by your voice, whats the difference? Seems innovative regardless of the tech underneath.
georgehillabout 1 year ago
Why did they upload the code to some random site instead of GitHub?<p>EDIT:<p>&gt; But let&#x27;s call a spade a spade – this is a blatant lie. And we&#x27;re about to expose it with the first partial release of the source code for its so-called &quot;large action model&quot;.<p>FYI, Text to Action is possible. I personally tested a couple of apps, but I don&#x27;t think anything reliable exists like we humans.<p>I would not disregard what they claim is completely false.
评论 #40135592 未加载
评论 #40135609 未加载
评论 #40135572 未加载
评论 #40135585 未加载
rideontimeabout 1 year ago
Taken down, now archived here: <a href="https:&#x2F;&#x2F;web.archive.org&#x2F;web&#x2F;20240423183114&#x2F;https:&#x2F;&#x2F;github.com&#x2F;rabbitscam&#x2F;rabbitr1" rel="nofollow">https:&#x2F;&#x2F;web.archive.org&#x2F;web&#x2F;20240423183114&#x2F;https:&#x2F;&#x2F;github.co...</a>
评论 #40164881 未加载
saltsamanabout 1 year ago
...Aaand It&#x27;s Gone!
评论 #40147966 未加载
anon115about 1 year ago
LOL
godelskiabout 1 year ago
Please upload source in an unzipped format. If the concern is about GitHub taking it down, use an alternative. There&#x27;s plenty and many other ways to distribute source in an uncompressed manner.<p>Otherwise this is indistinguishable from a hack. How do I know these zips are secure? The mega and pixeldrain report different sizes. Rabbit is entirely about hype and a scam, how are we supposed to know this isn&#x27;t the same nefarious ploy?<p>I appreciate what&#x27;s being done and think it&#x27;s good to call out these scams (I&#x27;ve done so myself) but help by building some trust. We understand the need for anonymity but a nefarious actor could just as easily mascaraed as the same repo. And if you do need files downloaded, provide hashes.<p>(Fwiw, xz, despite recent events, is great at compression and can help you reduce your bandwidth if needed)
评论 #40135715 未加载
评论 #40135778 未加载
评论 #40135873 未加载
评论 #40137924 未加载
评论 #40136328 未加载
评论 #40136162 未加载
评论 #40137263 未加载
评论 #40135938 未加载
rvzabout 1 year ago
Big if true.<p>Could set back the AI device hype 5 years back after Humane getting exposed as another scam.
评论 #40135665 未加载
paxysabout 1 year ago
Meh. I don&#x27;t think they ever hid the fact that the device is basically a ChatGPT wrapper. As long as it can achieve what it advertises, who cares how the backend looks? At least it has the decency to charge a reasonably price ($200, rather than $700 + subscription like the Ai Pin).
评论 #40135733 未加载
评论 #40135738 未加载
评论 #40135693 未加载
评论 #40135909 未加载
评论 #40135667 未加载