mseri

8 ms·

1.
Z1T: Sparse Transformer‑Like Models for Probabilistic Hardware(extropic.ai)
2 points by mseri 11d ago|0 comments
2.
by mseri 2mo ago
Ok, done: https://github.com/mseri/zunzuncito My main focus is systems with very low ram. On my M1 with 8Gb or RAM llama.cpp cannot run gemma4 26b, but this implementation works fine with 5-6 tok/s. I also replace
3.
by mseri 2mo ago
Nice! I had done the exact same with gemma4 26b, both for my Intel laptop and for my M1 with 8Gb RAM (with also q4 and turboquant). I don’t use it much since there are dumber but way faster models to run, but I should clean up the code and
4.
ZAYA1-8B: Frontier intelligence density, trained on AMD(zyphra.com)
5 points by mseri 4mo ago|1 comments
5.
by mseri 9mo ago
Finally! Great to hear. I had the exact same experience, but gave up at the moment of ID verification… too much hassle indeed
6.
by mseri 10mo ago
As much as I agree with the need for digital independence and the fact that universities (and governments) in Europe are over reliant on US tech, it is not as simple as you describe. There is a lot more happening in the administrative and i
7.
by mseri 10mo ago
Thanks indeed!
8.
by mseri 10mo ago
Thanks for the link. I had missed the other two submissions. If any admin is around, they should probably be merged. This is the other one: https://news.ycombinator.com/item?id=46055863
9.
EU Council approves Chat Control mandate for negotiation with Parliament(techradar.com)
164 points by mseri 10mo ago|159 comments
10.
Olmo 3: Charting a path through the model flow to lead open-source AI(allenai.org)
390 points by mseri 10mo ago|125 comments
11.
by mseri 11mo ago
Google has a great aid to reduce the attack surface: https://github.com/google-research/arxiv-latex-cleaner
12.
by mseri 11mo ago
I just want to say thanks to him! And very poor article by Politico
13.
LFM2-2.6B: Redefining Efficiency in Language Models(liquid.ai)
5 points by mseri 1y ago|0 comments
14.
Cognitive and AI scientists call to reject uncritical adoption of AI in academia(bloodinthemachine.com)
5 points by mseri 1y ago|0 comments
15.
MistralAI released a new Magistral Small 2509(huggingface.co)
8 points by mseri 1y ago|0 comments
16.
Granite docling 258M: a small multimodal model for efficient document conversion(huggingface.co)
2 points by mseri 1y ago|0 comments
17.
by mseri 1y ago
META did pirate basically all books in Anna’s archive but if I remember correctly they just whispered a a cried sorry and it ended up as that. Why are they also not asked to pay?
18.
by mseri 1y ago
True, but we should also remember that some services like the fast responses and the image generations (may?) run in US data centres also for Mistral. So that part of the data, in principle, may end up in the ends of other extra European co
19.
Apertus 8B and 70B – a new open multilingual LLM from Switzerland(actu.epfl.ch)
71 points by mseri 1y ago|4 comments
20.
by mseri 1y ago
Sounds all cool and interesting, however: > By submitting User Submissions through the Services, you hereby do and shall grant Inception a worldwide, non-exclusive, perpetual, royalty-free, fully paid, sublicensable and transferable lice
21.
HuggingChat is shutting down (for now)(huggingface.co)
2 points by mseri 1y ago|0 comments
22.
LLVM: InstCombine: A PR by Alex Gaynor and Claude Code(simonwillison.net)
3 points by mseri 1y ago|0 comments
23.
by mseri 1y ago
Some more details here: https://arstechnica.com/tech-policy/2025/06/openai-says-cour... And here are the links to the court irders and responses if you are curious: https://social.wildeboer.net&#x2
24.
by mseri 1y ago
Jim Portegies (TU/e, Netherlands) and Jelle Wemmenhobe have done a lot of research on this, using their “waterproof” (controlled natural language compiled to coa) to test this directly in class. The results are very interesting, and in
25.
by mseri 1y ago
It is a bit annoying that the article does not link any relevant research. There is a wikipedia page on the topic ( https://en.wikipedia.org/wiki/Coffee_ring_effect ), but afaik it is an interesting problem in many diffe
26.
Low-power 2D gate-all-around logics via epitaxial monolithic 3D integration(zmescience.com)
5 points by mseri 1y ago|0 comments
27.
by mseri 2y ago
It reliably fails also basic real analysis proofs, but I think this is not too surprising since those require a mix of logic and computation that is likely hard to just infer from statistical likelihood of tokens
28.
by mseri 2y ago
You can choose the quantization by appending the right tag to the model name, but they don't support other more advanced useful features (e.g. you need a special flag to enable flash attention and you cannot use KV cache quantization f
29.
MirageVPN and the discovery of two OpenVPN CVEs(blog.robur.coop)
1 points by mseri 2y ago|0 comments
30.
Released llamafile 0.8.13 with gemma2, new whisper and Stable Diffusion CLI(github.com)
2 points by mseri 2y ago|0 comments

More ›