Tepix

34 ms·

931.
by Tepix 2y ago
I'm so glad it's written in Rust! You know, for performance and safety and discussions at the coffee machine.
932.
by Tepix 2y ago
That really depends on the use case, does it not?
933.
by Tepix 2y ago
Tried it in mobile safari. When i i highlight a word and then use one of the font controls, i expect the word to remain highlighted. Also i can't seem to remove formatting like bold/italics by selecting it again.
934.
by Tepix 2y ago
When you train a neural net for Donkeycar with camera images plus the joystick commands of the driver, isn't that close to nerve impulses already?
935.
by Tepix 2y ago
The sentence "In addition there's a interest in having a lot of memory for LLM acceleration" was supposed to say "In addition there's a interest in having a lot of memory bandwidth for LLM acceleration" but i
936.
by Tepix 2y ago
Of course it's a competitor. Only a fraction of M3 Ultra sold will have 512GB RAM
937.
by Tepix 2y ago
You seem to disagree. Here's an interesting study where the researchers used an OpenAI-LLM-based tool to grade student papers and by grading them 10 times in a row, they got vastly different results: https://rainermuehlhoff.
938.
by Tepix 2y ago
You will be limited to a much smaller context size with half the RAM even if you're using a smaller model.
939.
by Tepix 2y ago
Running something like Qwq 32b q4 with a ~50k context will use up those 128GB with the large KV cache.
940.
by Tepix 2y ago
iGPUs have been getting ever closer to entry level and even mid-range GPUs. In addition there's a interest in having a lot of memory for LLM acceleration. I expect both CPUs to get more LLM acceleration capabilities and desktop pc memo
941.
by Tepix 2y ago
I think having a (small desktop) system with Strix Halo plus a GPU to accelerate prompt processing could be a good combo, avoiding the weakness of the Mac Ultra. The Strix Halo has 16 PCIe lanes.
942.
by Tepix 2y ago
Interesting to see the performance gain. But without file i/o and network access, what are the use cases?
943.
by Tepix 2y ago
I'd start with decent binoculars, you can see four jovian moons with them
944.
by Tepix 2y ago
Have you checked out the Le Robot project by Huggingface? You can build two robot arms SO-ARM100 (leader+follower) for less than 250€ or so. And you can train the model by yourself without relying on cloud services. Some URLs to get your st
945.
by Tepix 2y ago
I'm not in the loop, are 14 inch or larger mirrors affordable nowadays?
946.
by Tepix 2y ago
As a quick alternative, why not use a freelancer?
947.
by Tepix 2y ago
More consistent? That's not a given with LLMs unless you set the temperature to 0.
948.
by Tepix 2y ago
> Mark targeted the EFF, not a news outlet, in contrast. "Mark not only saw how it works, he had the documents to prove it. He brought us over a hundred pages of authenticated AT&T schematic diagrams and tables. Mark also sha
949.
by Tepix 2y ago
There are more recent estimates going from 6-20 trillion now: https://bigthink.com/starts-with-a-bang/galaxies-in-universe...
950.
by Tepix 2y ago
They don't call it "full self driving", that's a good start
951.
by Tepix 2y ago
If the user paid attention and it's a software issue that caused the crash, i don't see why not
952.
by Tepix 2y ago
The Dark Energy Spectroscopic Instrument (DESI) Legacy Imaging Survey has recorded more than one billion galaxies. But there are many more galaxies in the observable universe … somewhere between 2 and 20 trillion
953.
by Tepix 2y ago
I'm using Brave for those occasions
954.
by Tepix 2y ago
They are promising but they are priced very far beyond the 30€/GB of the used RTX 3090
955.
by Tepix 2y ago
Very cool to see two promising new LLMs on the same day (the other one being Reka Flash 3 21b) with open weights. Now, bring on those multimodal LLMs with voice input and output please!
956.
by Tepix 2y ago
As a european user, i think you should use the IP address of the user to determin the time zone and have an opening time for every time zone!
957.
by Tepix 2y ago
If you can't afford it, don't. But perhaps you will find alternatives for some of your purchases.
958.
by Tepix 2y ago
Performance is good enough for non-reasoning models even if they're FP8 or FP4. Check the phoronix article, the difference between the 3090 and 4090 is rather small. There's weight-only FP8 in vLLM on NVidia Ampere: https:/&
959.
by Tepix 2y ago
It used to be that for locally running GenAI, VRAM per dollar was king, so used NVidia RTX 3090 cards were the undisputed darlings of DYI LLM with 24GB for 600€-800€ or so. Sticking two of these in one PC isn't too difficult despite th
960.
by Tepix 2y ago
Related: FOSDEM 2025 talk "Build your own timeline algorithm" https://fosdem.org/2025/schedule/event/fosdem-2025-5601-buil... Only on Mastodon!

More ›