smcleod

70 ms·

841.
by smcleod 3y ago
Yeah I’ve noticed similar, you can even explicitly say, do not truncate your output, or do not take any shortcuts. Make sure you haven’t missed anything ensure there are no placeholders etc…. And it ignores all of them.
842.
by smcleod 3y ago
Only supports private / closed LLMs like OpenAI and Claud. People need to design for local LLM first, then for-profit providers.
843.
by smcleod 3y ago
It's not a service you're config changes to - it's the local client software on your machine. Even if it was the service - you're not reverse engineering, decompiling, hacking, disabling disassembling or even really modi
844.
by smcleod 3y ago
What on earth were Slack thinking when they rolled out this new UI - it is an absolute disaster! They took something that was OK and made it terrible. The worst thing from a functional point of view was hiding all your workspaces behind the
845.
by smcleod 3y ago
They don’t unless you explicitly choose to share it with a specific app. It’s in their T&C.
846.
by smcleod 3y ago
They are not equal, I don’t think USB4 even has DMA does it? USB4 devices can have a minimum of 20Gbps link speeds. In comparison, all Thunderbolt 4 devices have a 32Gpbs minimum requirement. USB4 has a 7.5W minimum power delivery whereas T
847.
by smcleod 3y ago
Nice idea, it has an incredibly narrow column for the content though - it makes it very hard to read and it's something like 80-90% waste screen real estate - @dndndnd check out https://i.imgur.com/XTx0jjc.png which I
848.
by smcleod 3y ago
I like the idea, but no Apple Music or standard m3u / playlist support?
849.
by smcleod 3y ago
Was that on Linux or macOS (intel)? I’ve ran them for years with both and they’ve been rock solid.
850.
by smcleod 3y ago
Thunderbolt is so much nicer than USB, I mean the specs sort of talk for themselves but it’s also far more flexible - you can basically think of it like an extension of the internals of your machine - rather than an external interface.
851.
by smcleod 3y ago
That is really very interesting, Thank you so much for sharing!
852.
by smcleod 3y ago
Hey thanks for the reply. I wanted to see if I could run it up at home. I’m generally avoiding most closed sourced ML/AI tools now.
853.
by smcleod 3y ago
Self hostable? Open Source?
854.
by smcleod 3y ago
Just trying out the official container image for self-hosting along side the VSCode extension - I've got to say I'm really impressed with the scaffolding especially for an early stage project. The web interface for the LLM server
855.
by smcleod 3y ago
I was thinking just the same! When I think of Github Actions I think of horrific inline YAML scripting, missing core functionality, clunky UI and frequent outages.
856.
by smcleod 3y ago
Yeah but I remember thinking to myself every few years that surely next year will be the year that base model machines start at 32/64/… GB - but alas, it’s nearly the end of 2023 and your average computer still seems stuck on a me
857.
by smcleod 3y ago
I always thought “First” and “Last” relate to the order of the names and was compatible with cultures that have self family, family self etc….?
858.
by smcleod 3y ago
Personally I find Vzf faster with less resource overhead than QEMU, but I've had some issues when running nodejs applications inside it (specifically when using the vz backend on colima).
859.
by smcleod 3y ago
So - the default macOS virtualisation?
860.
by smcleod 3y ago
Trying to regain some relevance in a world that has largely moved on from caring about it's core product and the added flop of the 'metaverse'. Without doing things like this they won't be able to compete with the large
861.
by smcleod 3y ago
Not OP either but yes - that is exactly what I want!
862.
by smcleod 3y ago
This is legitimately the only thing I've been asking for from Firefox the last <many> years, I don't really care about webgpu, usb device access would be nice as would fine grained user-controllable permissions for extension
863.
by smcleod 3y ago
Technically yes - if you have lots of ram you can use that and your CPU, as you say, the performance would be pretty poor, though, especially as it’s a toll where you want to tweak your responses quite frequently. I’ve been running and old
864.
by smcleod 3y ago
You’re making me envious. Before I went consulting that is all I knew, now there is only darkness.
865.
by smcleod 3y ago
Select “technical details” > Error - Page not found.
866.
by smcleod 3y ago
That doesn’t have to be a silo, you can have the capability rather than a role and agree on specs / contracts as to how your tools/apis will interact and work with the bigger picture or the product.
867.
by smcleod 3y ago
I have to say I’m pretty sick of scrum meetings and regular formal sprints. 80%+ of the meetings feel like a waste of time only adding context switching. You could argue that such teams / enterprises just aren’t doing it right and that
868.
by smcleod 3y ago
How is Pinterest even still a thing? I’ve never heard of anyone using it and for everyone I know with an Adblocker it’s the first thing to go.
869.
by smcleod 3y ago
Pivollama
870.
by smcleod 3y ago
You can already do that with GGUF/GGML models which allow you to split between CPU and GPU. Obviously there is a performance hit when running on your DDR5 and CPU compared to HBM/GDDR and GPU but it’s better than nothing.

More ›