prima-facie

5 ms·

1.
by prima-facie 5d ago
I use grm when I need worktrees, which makes admin a bit easier; https://github.com/hakoerber/git-repo-manager
2.
by prima-facie 11d ago
For one-shotted apps you still need to: 1. know how to do it well 2. maintain it 3. host it (if online)
3.
by prima-facie 15d ago
> What if the world had one GPU per person? It already does, in the form of mobile phones. It's just that these GPUs are not very powerful _yet_. But the building blocks to enable offline agents on mobile phones are already in place
4.
by prima-facie 27d ago
From the article: To block the scripts open the uBlock dashboard, select My filters, and add: ! AliExpress AWSC fingerprinting scripts ||assets.aliexpress-media.com/g/AWSC/uab/*/collina.js$script,domain=al
5.
by prima-facie 27d ago
This is not limited to Bluetooth in any way. In pavucontrol I can see Firefox outputting audio when on AliExpress even though nothing is playing. The uBlock filter fixed it.
6.
by prima-facie 27d ago
This is very cool, thanks
7.
by prima-facie 28d ago
You can manage these easily with the huggingface python app. hf download hf://unsloth/Qwen3.8-27B-GGUF \ Qwen3.8-27B-UD-Q4_K_XL.gguf and then see them with `hf cache ls`. Prune old versions with `hf cache prune`.
8.
by prima-facie 1mo ago
For SaaS forges there's also upd.dev launched recently. For self-hosted forges: forgejo is great but also cgit is pretty good for exporting repos. With the recent GitHub fiasco I've split my repos across multiple forges. Git was a
9.
by prima-facie 1mo ago
Silly question maybe: why use this at all when you already have a well known, decentralised, vendor-independent, almost real-time, and persistent notification system at your disposal? I'm talking about email of course.
10.
by prima-facie 1mo ago
> "Didn't you build this like... last week?" This captures the essence of AI generated work Vs your own work. AI generated work just doesn't seem to persist in my memory. If I design the application model myself then
11.
by prima-facie 1mo ago
I'm interested in this as well. I would assume all of them from the announcement. I.e. they are removing that functionality altogether from regular Gmail (not Workspace).
12.
by prima-facie 2mo ago
That's not how you're supposed to use a LLM. This is nonsense.
13.
by prima-facie 2mo ago
My point is that a generalist-humanoid robot can be used for nefarious purposes more easily than a specialised robot like a Roomba. A humanoid robot could, in an extreme case, perform a civil arrest or prevent you from leaving your house un
14.
by prima-facie 2mo ago
This is exactly my point, thank you.
15.
by prima-facie 2mo ago
Doing what's ethically right rarely means taking the most convenient path. Fridges and pantries take a considerable amount of space in our homes. These could become part of an automated storage system.
16.
by prima-facie 2mo ago
A lot of these household tasks could be automated without the need for humanoid robots. Roombas can clean your floor, automated rubbish disposal can be integrated into houses and flats, grocery delivery and storage could be done via an auto
17.
by prima-facie 2mo ago
I think this letters tells us everything we need to know about the man. I've rarely seen so much flattery towards the current administration, contradiction, deflection, half-truths and hypocrisy on a single page. This man is scared o
18.
by prima-facie 2mo ago
I found the following config to bring a great QoL improvement: FILE: .gitconfig ---------------- [alias] ci = commit fix = commit --fixup [rebase] autosquash = true Now you can do quick commits a
19.
by prima-facie 2mo ago
I understand this is important mainly for Nvidia GPUs. Is there any benefit at all for this (vs the existing VA-API) on Intel and AMD graphics? VA-API seems to work very well on both of these platforms, as far as I've tested.
20.
by prima-facie 2mo ago
Actually not really, Lunar Lake chips with Xe2 graphics have VVC decoding in hardware which can do 8K. https://en.wikipedia.org/wiki/Versatile_Video_Coding
21.
by prima-facie 2mo ago
Phones are already running models locally which can be used in the field for specific use cases. Maybe not for frontier coding just yet. Also you don't need to be connected to the network to use a local AI in many instances. If all mob
22.
by prima-facie 2mo ago
This piece is a bit all over the place. I immediately toggled off the `AI enhancements` and read the draft instead. The internet is already full of AI slop, I find human text a lot more valuable, even if unpolished. LLMs will not be central
23.
by prima-facie 2mo ago
There's a glimmer of hope with ROCmFP4 which seems to double the current throughput: https://github.com/charlie12345/rocmfp4-llama
24.
by prima-facie 2mo ago
The Strix Halo is a great dev machine and a mediocre AI machine. You can run Qwen 3.6 27B at a decent speed, or larger MoE models, and that's about it. For some that's more than enough though, myself included.
25.
by prima-facie 2mo ago
Sure but where do you stop spending? :) https://github.com/jamesob/local-llm
26.
GL.iNet Flint 4 Announced(gl-inet.com)
1 points by prima-facie 2mo ago|1 comments
27.
by prima-facie 2mo ago
MediaTek Quad-Core chip, 10G Wan/Lan, SFP+, 2G RAM.
28.
by prima-facie 2mo ago
> https://openwrt.org/voting/2025-02-12-openwrt-two Glinet are doing a great job with their routers. I have the Beryl AX which is fully openwrt compatible. The new Beryl 7 is also fully compatible now. Mediatek chip
29.
by prima-facie 2mo ago
At the moment, for around $4000-5000 you can either have speed (a GPU + 32GB VRAM), or you can have capacity - a DGX Spark/Halo, but not both. I think once someone comes up with a machine which has both it will easily sell for $10000 a
30.
by prima-facie 3mo ago
And this is exactly my point, the OEMs have more lobbying power and leverage. Anthropic might be valuated at whatever amount, but they're a new player and their only product is a piece of software - which others like Google, OpenAI, et

More ›