←
vtroc7396 posts
vtroc7 avatar

vtroc7

@vtroc7

ai hacker

◷ Joined January 2025▧ 93 Photos and videos
vtroc7 avatar
vtroc7@vtroc7·Nov 17

В оригинале "Tolerated by" но на русском это прямо жёстко звучит :'D

this is why you shouldn't use AI for translation LMAO especially landing pages it literally says "500k developers put up with us"

Post attachment
vtroc7 avatar
vtroc7@vtroc7·Nov 17

You can now get a decent, working 400k-line codebase for $200. A game, a website, a CLI tool, a driver... whatever But how much would you pay to actually understand those 400k lines? And is that understanding worth paying for at all??

vtroc7 avatar
vtroc7@vtroc7·Nov 17

SP support is coming to iw4l (slowly but surely) I also want to support mw3 survival and blops1 zombies even tho GSC might be incompatible with iw4

Post attachment
vtroc7 avatar
vtroc7@vtroc7·Nov 15

это очень смешно кто-нибудь, пожалуйста, сделайте раст компиляцию шустрее...

Jai compile times are so fast Claude can't believe it and has to double-check it's for real. When I asked it why it double-checked - it its sense of "normal" compile times comes from C++ and Rust. Even LLMs are amazed by what good engineering can achieve on modern hardware.

Post attachmentPost attachment
vtroc7 avatar
vtroc7@vtroc7·Nov 14

I want talaas to win eventually

The whole point of the original post is that burning the weights directly into silicon yields a chip capable of running Opus orders of magnitude more efficiently, at unit-prices orders of magnitude cheaper. That is what an ASIC is. If the only thing the system needs to do is run Opus 5.5, you can trim the fat required for general-purpose compute like Blackwell. It's already been done, too. Look up Taalas (Llama 3.1 8B). Granted, Opus is likely in the many-hundreds-of-billions class of parameters, so it'd still have to be a multi-chip module, but it would be nothing like the behemoth in your image. @grok explain more.

Daebak 앤디 avatar
⟳ vtroc7 retweeted
Daebak 앤디@MyDaebakCafe·Nov 13

I am actually disgusted by the timing of this drop. And this will get a bit more technical. Let’s do a real technical teardown of this "1,385-sigma Blackwell weight fingerprint" report that OpenAI and Anthropic are breathlessly promoting today, because banking on nobody reading past the headline is how DC narratives get manufactured. ​First, look at the actual CUDA maths: when PyTorch initialises model weights (torch.nn.init) with a fixed PRNG seed, curand/Philox parallelises random number generation across the GPU’s Streaming Multiprocessors (SMs). Because Hopper has 132 SMs and Blackwell has 148 or 152 SMs, the exact same seed lays out initial weights in a different spatial pattern depending on the GPU’s SM count. In huge MoE models, certain low-gradient null-spaces or unrouted slices preserve faint directional traces of those step-0 weights even after training. ​So yes, brute-forcing the seed and SM grid can show that a model’s step_0 weight initialisation script ran on a 148/152-SM Blackwell GPU. Now read what the summary explicitly admits in the fine print: 1. Initialisation is not full training: The report literally states that these signatures show how the weights were created, and "support (but do not prove) the inference that Blackwell also participated in model training". Initialising a model checkpoint takes seconds on a single rented dev node before loading the safetensors onto a standard Hopper cluster. The "2,000 to 3,000 Blackwell GPUs" and "$500 million" cluster numbers do not come from the weight fingerprint and Depue admits in his own thread that he simply "estimated" cluster sizes from release schedules and model details. ​2. Zero geolocation and zero proof of smuggling: The final sentence of the executive summary explicitly admits that the findings "do not establish the physical location of the chips, how they were acquired, or whether export control regulations were violated". Depue himself rented a cloud GB200 instance just to run his verification test. Commercial labs that are not on the Entity List renting bare-metal Blackwell nodes in offshore clouds (Singapore, Japan, or the Middle East) is completely legal under BIS rules and leaves the exact same SM fingerprint. Now ask yourself why a former OpenAI researcher dropped this on November 12, the exact day Jensen Huang was pushed out, with OpenAI and Anthropic immediately stepping in to "independently verify" it. US closed-source labs are watching their paid API moats get eaten alive by free Chinese open-weights models. Conflating a step-0 PRNG initialisation seed with a "$500M smuggled Blackwell cluster" hands DC hawks the exact pretext OpenAI and Anthropic want to regulate global cloud compute and kneecap open-source AI.

I have a major result to share. I’ve found conclusive forensic evidence that the top 4 Chinese AI labs—DeepSeek, Moonshot, MiniMax and Ant Group—have built their newest models on NVIDIA’s most advanced Blackwell chips, which are barred from export to China. This sounds insane, but I discovered that almost every AI model carries a hidden signature of the GPU that trained it. When a model is first set up, the starting values of its weights are laid out in a pattern that depends on the exact GPU. Each combination of GPU and seed number produces a unique pattern—a sort of fingerprint of the hardware used to create the model. Traces of that pattern survive training into the published model. Again, this evidence isn’t based on any insider information or leaking. It’s based on a forensic analysis of the model weights released by the companies themselves. I searched more than six trillion combinations of chip layout and seed number and matched them to open-weights models. In doing so, I found conclusive evidence that six Chinese AI models—including DeepSeek-V4-Pro, Kimi K3, and MiniMax-M3—were initialized on NVIDIA Blackwell systems. Four of the models match Grace Blackwell, the GB200 and GB300 parts in NVIDIA’s flagship liquid-cooled rack-scale systems. I then verified these results on real hardware. Running unmodified PyTorch with the seeds I’d recovered, a Blackwell B300 reproduced DeepSeek-V4-Flash’s untouched starting weights bit for bit, and a rented GB200 reproduced Kimi K3’s to a correlation of 0.999995. A Hopper H100 reproduced none of DeepSeek-V4-Flash’s, and changing the seed by one broke the match. (DeepSeek’s seed is 953049583 by the way.) The method also was verified to match 24 known open-weights models to the hardware that developers said they used. It is exceedingly unlikely that these results are a coincidence. Across six trillion combinations, the best accidental match should reach only about 8 sigma (i.e., eight standard deviations above what random chance would produce). The real matches range from 49 to 1,385 sigma, and even the weakest single test hits 20. These were data-center-scale training runs. With some analysis of release schedules and model details, I estimate that the largest, Kimi K3 and DeepSeek-V4-Pro, used 2,000 to 3,000 Blackwell GPUs each. Clusters of this size cost on the order of $150 to $200 million each to buy, and across these labs alone would add up to more than half a billion dollars of export-controlled GPUs, not counting experimental compute or otherwise. It’s quite likely that we see multi-billion dollar expenditure on export-controlled GPUs. I am highly confident in this result, and it has been independently verified by senior researchers at OpenAI and Anthropic. You can read the full report at unemploymentlabs.com/checkpo…. There is much more I’ve identified about these models through other forms of checkpoint forensics like this, and will have more to share soon.

Post attachment
vtroc7 avatar
vtroc7@vtroc7·Nov 11

escaping the gulag in CoD campaign was NOT easy t.me/contextrot

0:50
vtroc7 avatar
vtroc7@vtroc7·Nov 11

That's the only way I can finally hit a quad collat

more gameplay of Modern Warfare 2 with 512 client slots. ignore the typo... as Toru would say "chill on me bruh". 1024 slots is also done but I'm trying to fix the lag for that many.

2:07
vtroc7 avatar
vtroc7@vtroc7·Nov 9

been thinking about such an effect for a long time looks sick

last@last3740 Nov 9 0:12

0:12