与「Fringer」相关的搜索结果

#あきぴ#(@OkurimuSodaO)ちゃん コラボしたかったなぁ またの楽しみにするね #癖になるかわいさ# #推し事も地産地消# 🩷 #ナゴコス# #fujifilm_xseries# #Fringer#

Here’s the recap of @cz_binance’s fireside with Anthony Pompliano at Cantor Convergence in Miami: He showed up in Miami just weeks after the biggest hack in crypto history and still looked calm. No stress, no drama. Just there. What stood out: Trust gets built when things go wrong, not when everything is perfect. @cz_binance basically said resilience matters more than never getting hit. That’s how you actually move Bitcoin forward. He also talked about AI and regulation. “Any new technology will go through a phase we can’t make regulation ahead of time. It moves too fast.” Crypto already went through this. AI is going through it now. And security. The $6.3B hack showed the industry (and even banks) hasn’t been taking this seriously enough. AI is making attacks faster. The only real answer is fighting AI with AI and actually working together as an industry instead of everyone doing their own thing. No full recording out yet but @cz_binance said it should be somewhere. While most people are still panicking or pointing fingers, @cz_binance is still focused on the long game. Protecting users and building through the noise. That’s why a lot of us still trust him. 🔶

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 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.

BREAKING: 📰 Calvin Ridley says took truck keys back at Nashville’s airport after showing ID and photo proving ownership, Says valet accused him breaking finger. Ridley told reporters doesn’t know what happened and hadn’t heard from law enforcement as of Wednesday. What happens next? 🏈 #MHXbeyond# #NFL#

Introducing Token Exchange, a neutral, transparent and trusted marketplace connecting AI model service providers with customers. It gives customers one place to compare models side by side, review clear pricing and performance data, and start using them. For quality providers, it opens a new channel to reach more customers. Before listing, we check provider credentials and compare model fingerprints to help detect model swapping and inflated specs. After listing, we conduct regular spot checks and ongoing monitoring, and maintain a public credit record for each provider. One month in, the platform has more than 100 model resources, averaging over 50M calls and 700B tokens a day.

SOUTH KOREA'S "ANDROID TEST AREA" HAS 1 TEST: CAN YOU TELL SHE ISN'T ALIVE. THE ENGINEER CHECKS HER CHIN LIKE A DENTIST. THEN SHE SMILES AND HE FORGETS WHAT HE WAS CHECKING. Blue sign on the wall. Anatomy diagrams. A lab full of parts. Black hair. Cables down the neck. A glowing chip where the heart goes. He lifts her chin with 1 finger. Her eyes follow his. Cut. Blue hair now. Pearl earring. She laughs with every tooth showing. It's AI. @USUT in the corner. But the fantasy is specific. Not strength. Not labor. A face that reacts to being looked at. The chip in her chest glows blue. Nobody rendered a chip that does work. They rendered one that lights up when she's happy.