Search Articles — Sudonull

Search Results

In this project

Atlas 350: Ascend 950PR 3 times more powerful than H20 in FP4

https://sudonull.com/atlas-350-ascend-950pr-3-times-more-powerful-than-h20-in-fp4

Huawei Atlas 350 AI accelerator on Ascend 950PR delivers 1.56 PFLOPS FP4, surpasses Nvidia H20. Comparison of specs, HBM memory, price. For AI developers — details and nuances.

ROCm is catching up to CUDA in AI: OneROCm and Triton

https://sudonull.com/rocm-is-catching-up-to-cuda-in-ai-onerocm-and-triton

AMD unified ROCm into OneROCm for all accelerators. Triton writes kernels for AMD/Nvidia. vLLM inference on par with CUDA. Six-week releases. Test the stack on Strix Halo.

Cerebras IPO: profit, OpenAI and wafer-scale chips

https://sudonull.com/cerebras-ipo-profit-openai-and-wafer-scale-chips

Cerebras turned profitable and is preparing for IPO with an OpenAI contract for $20 billion. Analysis of architecture, risks, and prospects for the AI accelerators market.

From the web

Free Daily Horoscopes - Astrology Answers

https://www.astrologyanswers.com/horoscopes/

Each day includes unique surprises and revelations, so make sure you visit regularly to read your zodiac sign’s daily horoscope. Wondering what your horoscope for today (or even tomorrow) is? Sign up for …

Read Your Daily Horoscope! | Astrology Answers

https://www.astrologyanswers.com/

Read Astrology Answers' online daily horoscopes for your zodiac sign, Tarot readings, article content, Chinese astrology, numerology, & more!

网友调侃 Gemini 为北美大豆包,称其为最蠢 AI,为什么会被诟病难 …

https://www.zhihu.com/question/2047324900080051711

Jun 8, 2026 · 鉴定为没用过真正的豆包. 试了一次都感觉震惊, Gemini 至少不会胡说八道,豆包ai幻觉和胡编乱造的程度高的吓人. Gemini蠢但有逻辑,是合理的愚蠢,豆包完全就是42号混泥土拌意大利面 …

如何看Google发布的Gemini 2.5 Pro模型?

https://www.zhihu.com/question/1888290027890513537

Google发布Gemini 2.5 Pro: Gemini系列第一个2.5版本的模型,最高支持200万上下文,全模态输入,推理大模…

NLP 大模型与 CV 大模型的 “大”:同源而殊途的技术范式-云社区-华为云

https://bbs.huaweicloud.com/blogs/455174

Jun 27, 2025 · 随着多模态大模型(如 GPT-4V、Gemini)的发展,两者正走向技术融合 —— 例如,用语言模型的逻辑推理能力增强视觉理解,用视觉模型的空间表征能力补全语言的具象化缺失,最终共同 …

《记忆革命:Gemini 1.5如何让Transformer突破百万级上下文枷锁》

https://bbs.huaweicloud.com/blogs/453371

May 29, 2025 · 而Google DeepMind的Gemini 1.5却打破了这一宿命,首次实现百万级token的上下文处理能力,相当于让AI同时阅读《战争与和平》全集并记住每一个细节。 这一突破并非简单堆砌算力,而 …

Gemini Daily Horoscope - Astrology Answers

https://www.astrologyanswers.com/horoscopes/gemini-daily-horoscope/

16 hours ago · Read your free online Gemini daily horoscope for today! Use these expert astrology predictions and discover what your daily horoscope has in store.

如何评价谷歌最新发布的Gemini 2.5 pro模型? - 知乎

https://www.zhihu.com/question/1888168758947850201

这个时候,就可以给Gemini 2.5 pro文档,让Gemini针对自己正在写解决的问题「写一个教程」,然后不要用cursor的文档功能,而是直接把教程交给cursor 使用Claude 3.7 sonnet模型,基本上十有八九都 …

手把手教你在云上搭建本地大语言模型 - 华为云社区

https://bbs.huaweicloud.com/blogs/423514

Mar 10, 2024 · 比如大家都知道谷歌前阵子发布了Gemini,但可能有人不知道其还发布了开源的大模型Gemma,初次以外还有HuggingFace上也有很多出色的开源模型可供选择,比如Llama、Mistral等等 …

Gemini对香港和澳门开放了订阅版,这意味着什么? - 知乎

https://www.zhihu.com/question/2034415892650513007

而且Gemini的API接口我觉得也很难搞,去年我折腾几天,官方API接口基本不适配其他工具 (不知道现在咋样)。 我前一阵做了一个实验,用Gemini和国产的glm5.1对我的同一个工程进行改造,并且相互评 …

Trending Now