ismailmaj

6 ms·

31.
by ismailmaj 4mo ago
I had a much better experience gaming on MacOS than I ever did using Linux + X (it was first dual boot with Windows, then double GPU + VM, then proton). But it's mainly because I do not target anymore games with stringent anti-cheats o
32.
by ismailmaj 4mo ago
Gemini is a huge team while Gemma is relatively small. They can totally do this at a loss with no ulterior motive. They remind me a bit of HuggingFace, create something great then make money … maybe.
33.
by ismailmaj 5mo ago
I expected a 4090, maybe 2. I did not expect 8xH200 for a 2B model.
34.
by ismailmaj 5mo ago
I just asked the model details about the incoming spaceX IPO and it responded with “There’s no confirmed SpaceX IPO. Elon Musk has said for years that SpaceX itself won’t go public”. It took me two push backs and specifically asking for web
35.
by ismailmaj 5mo ago
3 things, they can, there is a precedent for that with Google v. Oracle for Java, and they have something! AMD engineered something called HIP which is CUDA API compatible libraries that targets AMD's hardware, it's the closest th
36.
by ismailmaj 5mo ago
Their moat is cuda and cuda libraries and everything built on top. When a new architecture drops, it's always PyTorch running on CUDA, other PyTorch backends are best effort, even if they reach feature parity, many industry power users
37.
by ismailmaj 5mo ago
I'm thinking that personally, technology is not bad in a vacuum and not necessarily bad in society, but it just reveals that our system is ill-equipped to guarantee good usage of it. We could have fun defining what's good usage bu
38.
by ismailmaj 6mo ago
Unclear if it's the only cause but wafer scale is great for very low latency, but loses to throughput per dollar compared to classic Nvidia like GPUs. I don't think they can reduce the gap, SRAM is just more expensive than HBM and
39.
by ismailmaj 6mo ago
Obsolete because of what? Because with limited hardware you’re never aiming for state of the art, and for fine-tuning, you don’t steer for too long anyway.
40.
by ismailmaj 7mo ago
I don't know why people mess with tesseract in 2026, attention-based OCRs (and more recently VLMs) outperformed any LSTM-based approach since at least 2020. My guess is that it's the entry-point to OCR and the internet is flooded
41.
by ismailmaj 7mo ago
Assuming 2 bit per values (first bit is sign and second bit is value). actv = A[_:1] & B[_:1] sign = A[_:0] ^ B[_:0] dot = pop_count(actv & !sign) - pop_count(actv & sign) It can probably be made more efficient by taking a colum
42.
by ismailmaj 7mo ago
Confusing, since this is specific to an architecture that no one making money will use (8B is consumer space, not enterprise). The produced code shouldn't hold much interesting IP?
43.
by ismailmaj 7mo ago
Oh no, cortisol spike in my text-only forum.
44.
by ismailmaj 7mo ago
The title and the repo uses 1-bit when it means 1.58 bits tertiary values, it doesn't change any of my arguments (still xors and pop_counts).
45.
by ismailmaj 7mo ago
You drop the memory throughput requirements because of the packed representation of bits so an FMA can become the bottleneck, and you bypass the problem of needing to upscale the bits to whatever FP the FMA instruction needs. typically for
46.
by ismailmaj 7mo ago
Ah yes, OpenAI the puppet of Microsoft that is currently declaring war against GitHub, sounds logical.
47.
by ismailmaj 7mo ago
Any place we can find the code?
48.
by ismailmaj 7mo ago
Don’t think that’s a fair interpretation of what I said. Liquid money rich? No. Can get pulled for big tech packages? Also no, for most of the employees. AFAIK, big tech didn’t aggressively poach OpenAI-like talent, they did spend 10M+ pay
49.
by ismailmaj 7mo ago
I don't see a world where they become threatening and the employees don't become rich from investors flooding in.
50.
by ismailmaj 7mo ago
In my experience, tracking objective things like "nutrition" and "sleep hours" is immensely useful to reflect on what went wrong, and tracking subjective things like "mood" or "stress" is useless give
51.
by ismailmaj 7mo ago
Surprised with the push back of the comments, getting effects on Rust would be a dream. Could even enable some stuff like passing loggers around not by parameters but by effect.
52.
by ismailmaj 7mo ago
Don’t need the EU for that, they are hitting everything in 2026 including unemployment, though nothing passed yet.
53.
by ismailmaj 7mo ago
I bought the M4 air given that the consensus was also that it was the best value for 1k, but I ended up returning it and went for pro base model for a few reasons (still valid for M5 AFAIK from a bit of research): - pro motion (120hz screen
54.
by ismailmaj 7mo ago
Semantic end of turn being 300ms of silence is horrible because I ended up intentionally um-ing to finish my thoughts before getting answer. It was difficult to detrain and that made me stop using voice chat with LLMs all together.
55.
by ismailmaj 8mo ago
3.1 feels to me like 3.0 but that takes a long time to think, it didn't feel like a leap in raw intelligence like 2.5 pro was.
56.
by ismailmaj 8mo ago
Cannot trust those damn Americans during on call.
57.
by ismailmaj 8mo ago
Warp specialization is an abomination that should be killed and I'm glad this could be an alternative. I hope they can minimize the bookkeeping costs because I don't see it gain traction in AI if it hurts big kernels performance.
58.
by ismailmaj 8mo ago
top 10 elo in codeforces is pretty absurd
59.
by ismailmaj 8mo ago
What I'm getting from you is that, given competence (and here it's a manager being able to make himself dispensable) things "work out". I tend to agree but I have a feeling that it takes a certain number of bad apples mi
60.
by ismailmaj 8mo ago
I see your point, this reminded me of the incompetent/malicious section. Yes, leadership could do the right thing, but the right thing is difficult to measure by shareholders and can only grasped by shareholders until it is too late (D

More ›