Grid: Free, local-first, browser-based 3D printing/CNC/laser slicer
(grid.space)393 points by cyrusradfar 3 days ago
393 points by cyrusradfar 3 days ago
(ob. discl., I work for a company which sells software in this space)
I wrote up a bit on Carbide Create at:
- https://willadams.gitbook.io/design-into-3d/2d-drawing (note that there is a link to a free (as in beer) download for Windows or Mac OS at that link)
- https://willadams.gitbook.io/design-into-3d/toolpaths
Other commercial programs which one licenses and installs and which don't intrude beyond that include:
- MeshCAM https://www.grzsoftware.com/
- Alibre https://www.alibre.com/ (note that there is a CAM option which is a re-badged MeshCAM)
- Moment of Inspiration 3D https://moi3d.com/ (this is probably the next commercial package I try)
and of course FreeCAD has a CAM Workbench which has seen great strides and Solvespace has a basic facility for G-code generation and some folks just program G-code/CAM directly --- I've been working on a tool for that myself: https://github.com/WillAdams/gcodepreview
Can it be locally installed in docker or something? It's kinda a bummer when I need to do something and there's a connection problem or the server is down.
Edit: looks like yes! https://github.com/GridSpace/grid-apps I will try it then.
In fact I had that only a month or 2 ago with fusion 360. Something in their cloud was down so I couldn't export to STL and i really needed that urgently.
Any circuit designers? looking to hobby, but what I saw was all proprietary
KiCAD has pretty awful UX though. I've tried all of the FOSS PCB design apps except LibrePCB (on my to-do list) and Horizon EDA is definitely the one I'd recommend (even though it also has a fair amount of UX oddities it's much better than KiCAD).
DesignSpark PCB is also decent - only minor UX mistakes like warping the mouse when you zoom.
- 100% free, no subscriptions, no accounts, no cloud
- Local-first: all slicing and toolpath generation runs on your machine
- Works in any browser, even offline once loaded
YES!
I think a new type of open source is emerging centering around what is now possible in browsers. Browsers have a great track record when running legacy projects. Relying on a backend could be a liability for longevity.
I built opal editor myself, a local first open source free markdown editor with these same principles, https://github.com/rbbydotdev/opal
> Works in any browser, even offline once loaded
That, my friend, is not how offline works. You will be required to have internet access in one way or another. Offline works 100% locally no matter if you have internet or not.
You're misunderstanding how websites work. You can install it locally on your computer and run it via localhost https://github.com/GridSpace/grid-apps
There's an electron build, although why you'd use it over a native slicer is beyond me.
With a stand-alone application once you download it in your file system you know exactly where it is and how to create backups etc.
A "browser cache" is just an opaque bit of storage. What if you need to update/reinstall your browser or want to switch? I wouldn't trust important data to it.
I generally feel uncomfortable how so many applications are browser-only these days. The thought of having important data in a tab that you might close by mistake at any moment is uncomfortable. Browsers should really only be used for fleeting content, not productive work.
That used to be the case, but there’s a new API worth checking out, you can read and write files right on your computer https://developer.mozilla.org/en-US/docs/Web/API/File_System...
I've used kiri:moto for several simple CNC projects!
This probably won't scroll to the correct place on the page but there's some images of my project at https://hcc.haus/propmania/#2024-palm-torches and https://static.cloudygo.com/static/Prop%20Making/2024%20Palm...
I used it instead of the terrible closed source Easel App for a CARVEY hobby CNC. For metal milling I find Fusion 360 is necessary.
Curious if you can elaborate on what's missing or failing, to require Fusion 360?
no shared lineage. Cura and Kiri started around the same time (2011/2012), but as completely separate projects. Cura is a C++ desktop app and Kiri has always been 100% browser-based (no cloud, all computation in the browser sandbox). the licenses are different, too. Cura/Prusa/Orca are GPL based and Kiri is MIT.
OT: Why is that Alphabet, Mozilla, Apple, etc can get together to create web standards that allow anyone to create software that works cross-platform - only a browser is needed, but Microsoft, Alphabet, Apple, Canonical, etc can't get together to create standards that allow anyone to create software that works cross-platform?
You answered the question yourself: There is already a standard that allows anyone to create software that works cross-platform: the browser.
The browser is an extremely poor medium to deliver applications. It works, but barely, is a huge resource hog, fragile and it breaks way too often due to a lack of backwards compatibility between browser versions of the same manufacturer. I have a small app that I support and it's been fun to get it to work in the browser (instant cross platform support was indeed the driver) but the experience is still sub-par compared to what I could do on a local application.
Unfortunately, I think all these things are externalities - or at least, areas that don't impact revenue enough to get companies to change.
I too wish that software would be efficient, robust and long-lasting. But it seems that most people don't care about this enough (compared to other factors) to force change. (Alternatively, they are locked in to platforms they don't like to use.)
Apple make money from the App Store and from selling their hardware, so why should they want to invest on something that let people install software bypassing the App Store or that works on other platforms?
Alphabet make money from ads, so they want web pages, apps on Android and Chrome everywhere.
Mozilla make money from Google.
Microsoft make money from software licenses and subscriptions and from cloud services. They might be interested in cross platform installation.
At the moment what we have is PWA and WASM and icons on the desktop.
There are many projects that try to make cross-platform mobile apps easier, including Google's own Flutter. I haven't heard of them getting much cooperation from the teams working on Android or iOS, though.
At least for stuff that doesn't use device API's much, it seems like websites are the way to go. They're a whole lot easier to build than mobile apps.
The API surface becomes the lowest common denominator of all the platforms it supports, possibly with a path to support platform-native features, but probably in a way that’s necessarily not as good as native.
I think we already have plenty of avenue in ‘solutions’ like Electron to let people build bad apps.
Ah, I'm always up for a tangent.
The boring answer from Capt. Obvious. Incentive alignment.
That said, WebAssembly might be the trojan horse. While it started as a browser compile target, WebAssembly System Interface (WASI) is extending it beyond browsers into filesystem, networking, etc. etc. etc.
Fingers crossed, we may get cross-platform standards by accident.
Simple. It is not in their interest to do this. It is a lot of work, for no revenue.
Am I weird in not being too surprised? It don't have experience with wire EDM but every toolpath generator or slicer I've ever used was just local software.
Truly. The slicer is able to generate bugs I've never seen before, in around 6 years of printing with several slicers and firmwares. Cura, Flashprint, Orca, Prusa, using Marlin, Sailfish, Klipper. None of them produced the weird stuff I find with Bambu's pipeline.
When the bugs don't creep up it's absolutely incredible, though.
For those wondering why having a browser based slicer is useful: teaching. The site mentions this, but I'll add my own experience that having good in-browser software like this is incredibly useful when you have a classroom full of students who a) aren't used to installing desktop software, b) are running a bunch of different operating systems (including chrome os), and c) have firewalls prevents them from installing local software anyway.
I wouldn't want online tools to be come the default (like google docs) but having them as an options is great. (I find onshape and photopea useful in this way as well).
A fun thing you can do with this excellent sw is to slice a 3d object into slices to cut with a laser cutter. Ie you'll get a bunch of layers of eg cardboard or plywood, which you can assemble into a large object. Increase layer height to thicker than your material to create gaps in between. This operation is the basis for some very nice looking creative stuff you can find on etsy or even high-end wood working stuff.
Part of me wants to be wary. The useful life of industrial machinery such as CNC mills is much longer than the lifespan of websites, so locally-installed software you own is usually a better choice.
But another part of me realizes that everyone is using Fusion360, despite the fact they have a history of taking away features to force people to migrate to paid tiers. So it probably doesn't matter.
> much longer than the lifespan of websites
But browsers (and browser technologies) have documented track of being fully backward compatible up to the beginnings of WWW, and it's not going to change.
Which actually is much much better than any other environment you can imagine - unless of course you use (and want to use) that one frozen in time 25 year old PC. And pray nothing breaks (y2k bugs and whatnot).
If the software is open source (and works offline) you can have it functional in 10 or 20 more years. And it will be "locally-installed software you own" you want.
> But browsers (and browser technologies) have documented track of being fully backward compatible up to the beginnings of WWW, and it's not going to change.
That can however be undermined if web apps are poorly built and depend on quirks and behaviors specific to a particular engine (or in some cases, even particular versions of a particular engine) in order to function.
So I would say this benefit applies specifically to web apps that thoroughly apply KISS — that is, using only the most boring, solidified, widely supported APIs and favoring robustness over bells and whistles — and make a point of testing against all three major engines. Those apps will likely stand the test of time and run even under future new engines. On the other hand, the ones with severe shiny API syndrome that only ever get tested against the latest Chrome are probably much more brittle and more likely to be broken N years after abandonment.
"Fully backwards compatible" isn't really true, and even if it were, then you're stuck using browser-based software and its myriad of inherent downsides.
People (generally) use web-based apps that are good enough in spite of the web stack -- not because of it.
I have a CNC mill made in 2006. It's still perfectly fine. It should still be fine in 2036. The most significant threat to its existence is the compatibility of OS drivers and software support in CAM tools. That and USB ports getting replaced by something else, which was a problem for earlier-generation machines that used RS-232.
Worst case - you could at some point rip out the brains and replace them.
CNC machines are somewhat basic machines really.
This looks great. I was hoping it would have been a good OrcaSlicer replacement for my FDM printer, but unfortunately it didn't generate any top surfaces (except for the topmost one) for a model I imported in. I didn't know if it was the printer profile (Creality.Ender3) or something else, but it seems I'm still using OrcaSlicer for the time being.
It's a shame they don't have an actual application for a truly offline experience. If they had both, people could have their cake and eat it too.
there are desktop builds https://grid.space/downloads.html that run entirely offline. you can also use the center menu to "install" it as a progressive web app.
So is this the software part? Or do they sell hardware too? Are there recommended ones?
I bought a bambu p1s recently and it can be used entirely offline.
You can import models to orcaslicer (open source), do your slicing, and export the g code file to SD card.
If you want to skip the SD card, block the printer's mac/ip address at the firewall and set up WiFi. Then send the print directly from orcaslicer.
That being said, my gut says bambu is going to slowly require a persistent connection to the cloud at some point. Maybe they think they are an EV car company.
Yes, you can use the P1S offline, but they've done everything possible to make it hard: crappy micro sd interface without the ability to use folders, USB plug absolutely inaccessible (see if you can even find it) and then, once you've found it it turns out that there is absolutely no way to use it to print with. They push you towards their closed source plugin 'for your benefit'. Fuck that stuff. There is absolutely no way I'm going to run a Chinese built piece of software that I can not inspect on my desktop. FOSS or bust.
I'm curious - which firmware version are you on? Just yesterday I got confused when I accidentally entered the "cache" folder (Handy sliced models?) on my P1S, so it looks like my machine definitely supports folders at least. I am located in the UK and bought it late 2025.
AFAIK, the hidden internal USB port is intended to charge a phone/other device used as a "touch controller".
I don't like their closed-source control and the way they've managed local integrations, but I also want to accelerate making things with a printer that just "prints" (akin to how some will buy into the Apple ecosystem for something that "just works"). I've now sold my previous open-source Klipper/Mainsail printer which I heavily modified.
If/when Bambu makes their ecosystem unbearable for me, I will switch away (probably to a Prusa). For now, the ease of use of the machine has led to much more things being made as I can trust it to work accurately and reliably.
Yes, this is the dilemma right now. I have 30+ prusa's and 43 Bambu's and another dozen or so K1s. They all have their quirks, weaknesses and strong sides but for sheer reliability and volume you can't beat the Bambus. But for development work I'd much rather have the oldest bedslinger that I have because it is so incredibly precise, even if it is terribly slow as well. K1 I would not recommend for anything, too fragile and too irregular from one machine to another, each is a different printer. To get to that level of lack of consistency must have taken some real thinking.
I'm not so sure. You've had people insisting that, or how they'll lock it to only their own NFC tagged filaments but lets be real, if they start imposing serious restrictions most purchases would go to other makers.
They make solid enough devices sure, but they dont exactly have a moat that would keep users buying their stuff if it were to start getting locked down. They have far more to lose than they have to gain doing so.
Kinda funny how things have progressed...
I work with the Smoothieware project. The V1 Smoothieboard was one of the first with ethernet onboard (although kinda borked). First thing that was advised to everyone was "never connect this to anything outside your local network"
Nowadays...it seems that warning has been lost. Even in the face of firmware updates that caused physical damage.
Something to be said for building your own printer.
Elegoo printers can be offline - you can run everything from the machine itself, as long as you have your model/s on a thumb drive. Or is that not what you mean?
They’re trying to introduce legislation that would require 3D printers to be online so that if you try to print a firearm, it won’t let you…
Granted, today, you can print offline.
Tomorrow? A firmware update might just brick it the next time it goes online or won’t be able to read the grbl
These people are so ridiculous. It'll fail on 1A and 2A grounds, not to mention challenges implicit from 4A and 5A considerations. They can't ban arbitrary information, even dangerous information, and there's a presumption of regularity - you're presumed innocent of wrongdoing absent evidence, so they can't legislate the assumption of criminality by default. They can't ban private creation of firearms and weapons, so long as other aspects of the law are being followed. They can't assert control over private property and mandate being online, this is equivalent to a warrantless search of private home activity. Arbitrary compliance costs and increased prices can amount to violations of 5A takings clause, and you can't bake in a violation of your right to refuse to incriminate yourself, especially with the vague, subjective nature of the proposed legislation. There's also 5A due process concerns, with the legislation being overbroad and arbitrary. 14A presents equal protections and lays the basis for discrimination between hobbyists and manufacturers and interstate commerce concerns.
The whole notion is about as anti-American and authoritarian as laws get, I don't see it as anything more than political grandstanding, and even if Washington passes it with statewide, unanimous endorsement, it won't last a year before 9th circuit court strikes it down on purely 2A grounds.
This reminds me of William Gibson's "The Peripheral" in which the protagonist runs a rural 3D print shop where everything has to be licensed and government approved. We live in the near future.
How would it know what is a firearm and what isn't? Seems trivial to defeat for someone who knows CAD, no?
And work really well until you have more than 20 files or so after which it becomes impossible to manage the SD. Your best bet here is to have a stack of SDs of what you'd normally put in folders. Really nice, especially given how easy it is to write on the outside of a micro-sd what the contents are.
Surprised this hasn't been shared here before.
Built by my former colleague, Stewart Allen (Co-Founder/CTO of WebMethods, CTO of AddThis, Co-Founder/CPO of IonQ, et al.).
What caught my attention:
- 100% free, no subscriptions, no accounts, no cloud
- Local-first: all slicing and toolpath generation runs on your machine
- Works in any browser, even offline once loaded
- Supports FDM/SLA, CNC milling, laser cutting, wire EDM
- Fully open source: github.com/GridSpace/grid-apps
Refreshing to see a tool that isn't trying to lock you into a subscription or harvest your data.