I was rather commenting on the tone that Thoughtworks decided to "take 10 engineers and put them in one room". I took that wording literally, imagining how they grabbed these 10 engineers with a big hand and dropped them off where they wanted them to develop their airline IROps system.
Setting up custom GitLab runners can be cumbersome in the beginning! We faced that multiple times with various customer projects for which we were using GitLab.com managed repos. That's why my colleague and I recently built a "GitLab Runners as a Service" [0] -- simply use your self-managed GitLab or GitLab.com account to login, adding pipeline runners is a 1-click operation! Under the hood we provision a Hetzner machine for you runners and automatically connect your GitLab group or project to it. Happy to hear your feedback!
This is pretty cool and one of the few things I couldn't find a viable alternative to in-house hosting so far. The only thing your "GitLab Runners as a Service" is missing is ISO 27001 (and optionally SOC 2) certification. That makes it again easier for your customer to get / maintain their ISO 27001.
Ha, let us look into ISO 27001 certification and see what it takes to obtain it (and SOC 2). Thank you for this valuable pointer! If you could just sign in (accounts are free and you can even try us out for the first 48h and not be charged), then we could message you to your email address once we have an update here.
I clicked 'Get started -> Login with GitLab.com' but for my tastes it wants way too much access just to add a runner:
> Grants complete read/write access to the API, including all groups and projects, the container registry, the dependency proxy, and the package registry.
> it wants way too much access just to add a runner
We wanted to make the onboarding as simple as possible so that we can:
0. Log you in through your existing GitLab.com or GitLab self-hosted account
1. List all your groups and projects to decide where to install the runner
2. Disable GitLab.com's own instance runners (otherwise our Runners are not picked up)
3. Install the Runner
4. Optionally: When installing an agent runner, we are able to reply to read and reply to all your issues where you tag us via `@Rocket` -- this runs completely on your machine that we host for you! It doesn't run on our infrastructure and we only access your runner machine for auto-cleaning its caches.
That's really all the RocketRunner does with you granting us permissions when logging in.
We are looking into ways how to possible scope access better! Thanks for your feedback!
I find the gitlab runner registration process a lot simpler than the github process, and a lot more flexible when you are self hosting the gitlab instance. We used kubernetes runners for both, and gitlab was clearly better.
The worst part about the github runner setup is that there is no built in support for using your own cache if you are using github.com and want to run your own runners. In order to use your own cache store for your runner jobs, you have to patch the runner image because the cache location is hard coded. Gitlab lets you choose your cache location as a standard feature.
Right! That's why running your owners runners is so valuable as you've got full control over the cache location which clearly makes subsequent pipeline runs super fast.
Thank you for building this, looks like it could be really helpful. We've been running our self-hosted GitLab with almost 99.99% uptime and a few runners but the configuration and update process has been a bit cumbersome. Will take a look at rocket runner.
99.99% uptime is impressive! So far our runners are likewise very reliable! It's just a Hetzner machine with our stack on top after all. Thanks for trying us out and please drop us an in-app message for any questions or comments!
Is it? Especially if you go the K8S route, setting up a runner infrastructure can be done within a day. Based on my experience, there isn't a lot of maintenance involved. It's mostly a smooth experience.
> React Flow examples show possible approaches, not production coverage. They omit edge cases, performance considerations, and maintenance paths. Overflow gives you the end result: hardened behavior, documented APIs and types, and components that are ready to ship.
So it's "us" vs. "they"? "They" actually build a very widely adopted open-source solution, why don't you contribute back if "they" omit so much?
Practically anybody actually creating with this class of models (diffusion based mostly) is using ComfyUI.
Community takes care of quantization, repackaging into gguf (most popular) and even speed optimizing (lighting loras, layers skip). It's quite extensive
I personally find nothing about ComfyUI to live up to that name. Node-based workflows are unruly, and you have to know in advance what you need to do for anything to work. Just poking around and figuring stuff out is nearly impossible even for technically literate but AI-inexperienced folks.
You could argue that is what pre-made workflows are for, but that doesn't work super well for users that are off the blessed path in terms of not having Nvidia hardware like everyone assumes. I personally find using stable-diffusion.cpp on the command line to be considerably easier to figure out. Last I saw, it's even shipping a usable demo web ui for those that really want one (my workflow benefits from heavier scripting, so point and click is far too slow and clunky).
LMStudio is a low barrier to entry for LLMs, for sure. The lowest. Good software!
Other people gave you the right answer, ComfyUI. I’ll give you the more important why and how…
There is a huge effort of people to do everything but Comfy because of its intimidating barrier. It’s not that bad. Learn it once and be done. You won’t have to keep learning UI of the week endlessly.
The how, go to civitai. Find an image you like, drag and drop it into comfy. If it has a workflow attached, it will show you. Install any missing nodes they used. Click the loaders to point to your models instead of their models. Hit run and get the same or a similar image. You don’t need to know what any of the things do yet.
If for some reason that just does not work for you… Swarm UI, is a front end too comfy. You can change things and it will show you on the comfy side what they’re doing. It’s a gateway drug to learning comfy.
EDIT: most important thing no one will tell you out right… DO NOT FOR ANY REASON try and skip the VENV or miniconda virtual environment when using comfy! You must make a new and clean setup. You will never get the right python, torch, diffusers, driver, on your system install.
Everything keeps changing so quickly, I basically have my own Python HTTP server with a unified JSON interface, then that can be routed to any of the impls/*.py files for the actual generation, then I have of those per implementation/architecture basically. Mostly using `diffusers` for the inference, which isn't the fastest, but tends to have the new model architectures much sooner than everyone else.
I encourage everyone to at least try ComfyUI. It's come a long way in terms of user-friendliness particularly with all of the built-in Templates you can use.
FWIW you can use non-sd models in ComfyUI too, the ecosystem is pretty huge and supports most of the "mainstream" models, not only the stable diffusion ones, even video models and more too.
Koboldcpp has built in support for image models. Model search and download, one executable to run, UI, OpenAI API endpoint, llama.cpp endpoint, highly configurable. If you want to get up and running instantly, just pick a kcppt file and open that and it will download everything you need and load it for you.
Yeah, I'm guessing they were bound to leave behind the whole "Get up and running with large language models" mission sooner or later, which was their initial focus, as investors after 2-3 years start making you to start thinking about expansion and earning back the money.
Sad state of affairs and seems they're enshittifying quicker than expected, but was always a question of when, not if.
National sovereignty is not national pride. Paying company taxes to the 28th regime evicts tax contributions to the company's home nation. Massive German tax money is already today used for non-national interests. EU-inc makes this only worse.
I don't understand the tax objection. The 28th regime changes corporate law, not fiscal law. A Berlin-based "EU-Inc" still pays full taxes (Körperschaftsteuer, Gewerbesteuer, etc.) to the local Finanzamt. It isn't a disembodied tax haven - it just standardizes the legal wrapper.
You are right that this threatens the status quo, though. If this works, founders based in Germany will likely abandon the GmbH. That will require swallowing some national pride and admitting that the current system is simply a less efficient, less competitive legal form for high-growth companies.
One thing I think is also worth mentioning are labor rights. I am not arguing against the German model of employee protection. Mitbestimmung could be viewed as a good thing, even if it will mean less power to the VC and / or founders. And frankly, I don't care if the consensus forces strict, German-style Mitbestimmung on the EU-Inc. Stricter form of EU-INC is still vastly better than nothing at all.
This is wrong on so many levels. The German economy profits _a lot_ from the membership.
You can't just look at the monetary in/out flows to/from EU accounts but to the system as a whole.
For example, decades of pegging the value of the currency down, thus facilitating exports on a massive scale and thus pushing up the performance of whole industries, without other member states that would have had their currency devaluated being able to compete on prize.
They probably didn't list any because estimating the upsides versus the downsides is what it's about, and that's very hard to quantify. Suggesting, like you seem to, that there are no upsides to the German economy whatsoever, is just not a serious argument so doesn't actually deserve a serious answer.
I agree with your characterisation of what is going on, and at some point, the EU states will have to decide for full fiscal integration or for removing the common currency. You can't have a common currency without a common fiscal union. So we either have to integrate more or desintegrate more, this inbetween we have now is not working very well. Speaking as a European, not sure what is better.
Not related to the comment, but in general I agree with you.
You can't have a single monetary system without complete unification, including tax systems, budgeting systems, governance models, retirement systems, benefits. I mean, you can, like we have now, but it's not sustainable, and eventually we all have it worse.
As a European, I would not want to go that way, since I'm afraid such a unified EU will be a bureaucratic monster that is even more centralized than the USA, and way more autocratic than any current EU state.
I'd rather take a step back, dissolve much of the EU's competences, and go back to pure trade union, dissolve the EURO as a currency, and let every member state take sovereign decisions on their own.
Even more centralized than the USA? That must be an outsider's view. You are aware that each state has its own tax laws? Corporate laws? Even criminal laws and traffic laws? And so on? Obviously there is federal law, too, but that's more of an umbrella covering all states in addition to local laws. The USA is a federation, just like Germany, Switzerland, or Austria are federations, and one common complaint you get out of these nations all the time is that they are not centralized enough!
The EU, on the other hand, is not a federation, it is a association of nations trying to figure out how to go together, be it as a federation or something different, because in today's world you need size and power to survive. Any EU state alone would be insignificant on the world stage and being sidelined one way or the other. The most powerful inside the EU would suffer the most, e.g. Germany or France, because while the EU is second or third on many metrics, the individual nations are small. Depending on which ranking you look at, the EU is just behind the USA in military and economic power, sometimes it is behind China in economy, too.
The strongest individual economic power of the EU is Germany, which globally usually comes in 4th place after the USA, China, and the EU as a whole. But this is deceiving, when you look at the absolute numbers. Germany's output is impressive for its size, but is still only about a quarter of China's! Germany alone is a dwarf and without the EU would be inconsequential. The post-Brexit UK is just learning this the hard way and all the sovereignity does not matter a damn. The top leadership of EU nations are not stupid, even if they could explain their reasoning better, and thus try to keep the EU going, despite its many flaws.
I recently became much more pro-total-unification, so let me give you this counterpoint: individually, European nations are no match to the major superpowers, neither economically nor militarily. We'll get gutted by divide-and-conquer approach. In contrast, bound much closer together (particularly with some form of pan-european armed forces), the EU would become a proper global superpower and a counterbalance for the USA and China.
There is not a single example of a multiethnic or culturally diverse empire or state that was successful after the rise of nation-states at the end of the 19th century.
All of them crumbled either peacefully or in bloody wars.
Why do you think such a bureaucratic monster that no one really wants would be an exception?
Maybe we can try it again in the 24th century, when the humans evolve enough, but for now, your best bet for a successful country is either an ethnonational or religion-national state.
You cant really believe any of the EU countries actually want to work together? The EU was only possible with the premise of countries keeping their autonomy. Believe me when I say that us Germans would rather go to war than merge with France, just as an example.
>> - yes, German taxpayers are indeed funding bicycle lanes in foreign countries
Just for people to know this, we are talking about 44Mn€ (20Mn€ + 24Mn€) of money in Peru as the commentator mentions below. It is wasteful but it has practically no effect on Germany. There are 1000x things that Germany does within Germany that moves the needle more than the talking about this rounding error. But that requires introspection and ownership of responsibility - both alien to the country’s normal compliance attitude of working.
That's the same fallacy as claiming that rich people save money by donating to charity.
Simple example:
- Germany gives 100 € to other EU states
- Those states 100 € to buy made in Germany goods
- German companies have 10 € profits from those sales
- German state collects 5 € from those profits
In summary, German taxpayers paid 100 € so the state can collect 5 € taxes, and companies another 5 €. Not a very good deal.
And this calculation is crazy optimistic, because it assumes that all money that Germany gives to other countries will be used to purchase German good. In reality, they will be spent willy-nilly and German companies may not see even 1 € in return.
This calculation is flawed. German state collects much more than 5€, because there is VAT etc and, of course, employees of that company pay income tax from their share of that 100€ they just got back.
What people like you dont understand is what we get back from investing into foreign countries. We dont just build bicycle lanes there because we are such nice people.
This can happen without putting the most part of intensive payload in a form of external dependencies that can be cut at any point by many factors, while losing all the skills and knowledge on how to effectively plant and harvest. Qualified farmers don’t appear magically over night, let alone people with the passion required to invest their own life into it.
And exporting the model that ravaged one own land, selling the poison that was forbidden in the countries exporting it, what kind of behavior is that? Letting some land having a bit of rest while the same depletion process is applied elsewhere, how fair is that?
You can do Paris-Berlin-Moscow-Beijing by train via the trans-Siberian, or via Ulan-Bataar with the trans-Mongolian.
Quite an iconic route that became much simpler administratively to travel in 2000s but perhaps again trickier now because of the situation with Russia.
reply