搜索“xAI”

on. In a later section, we will provide a deeper technical dive into the methodology, replay harness, and dataset. When looking at inference coding performance, OpenAI, Anthropic, xAI, and other frontier labs focus on three things. They look at performance per dollar versus interactivity (TPOT), TTFT (time to first token), and overall end-to-end task completion. Performance per megawatt is also important, considering that terrestrial datacenter power is a critical constraint (money is a social c

匹配字段正文

ata - everyone I’ve met with, emailed, 4:01 PM · Aug 19, 2026 · 92K Views 187 Replies · 82 Reposts · 1.92K Likes Nikunj Kothari @nikunj the models have no moat (OpenAI, Anthropic, XAI) the IDEs have no moat (Cursor, Windsurf) the harnesses have no moat (Cognition, Factory, LangChain) the app builders have no moat (Replit, Lovable, Bolt) the wrappers have no moat (Harvey, Abridge, OpenEvidence) the inference 10:57 PM · Aug 17, 2026 · 452K Views 349 Replies · 233 Reposts · 3.4K Likes Turner Novak

匹配字段正文

labs in adoption decisions at the frontier. To put it simply – the Chinese labs use all the time that American labs do pre-release testing to keep hillclimbing on benchmarks (SpaceXAI is likely far closer to the Chinese labs here). With the pace of progress being so fast, this is likely the largest determining factor of why Chinese labs stay at the frontier. This, so far, has been economically acceptable for the American labs, as they’ve still had massive demand for their models. As model self-i

匹配字段正文

tier and trailing models across domains? This question applies to open vs. closed weights, the US vs. China, and even the US frontier (OpenAI, Anthropic) vs. the US near-frontier (xAI, Meta). All of these questions have an economic valence: can the leading developers capture enough value vs. their competition to sustain the frenetic infrastructure build-out? The US/China framing has additional geopolitical implications. One way closed-weight developers could capture this value is if the open- vs

匹配字段正文

ech titans here are spending the most money. Interestingly Supermicro had one customer who alone accounted for 28 percent of revenues in fiscal 2026. This is almost certainly SpacexAI, with the Colossus 2 supercomputer, which is mostly being rented by Google and Anthropic, for an estimated $920 million and $1.25 billion per month, respectively. Supermicro also said that it had nine customers who spent over $1 billion. So that is CoreWeave and seven other companies. Those nine customers are aroun

匹配字段正文

ss Index™ in the last ten days went straight to the top of the leaderboard. State of the art is now measured in weeks. A month ago we published the first Humanness Index™ results. xAI's Grok TTS led the models at 95, five points under the real human baseline, and the story was how close synthetic voices had gotten to a real person. Those standings held for about three weeks. Then they stopped holding. On July 24, we added Speechify's Simba 3.2 to the arena, and it went straight to the top of the

匹配字段正文

ith tree-structured sessions that allow branching and navigation without losing context; Broad provider support via BYOK: Pi connects to Anthropic, OpenAI, DeepSeek, Gemini, Groq, xAI, OpenRouter, and Kimi through your own API key or existing subscription; Small core: Pi keeps its built-in orchestration and tool surface minimal, which can make it more straightforward to inspect and modify than larger agent frameworks; No governance layer: Pi has no plan mode, sub-agents, MCP, or org-level contro

匹配字段正文

tary information. The state of affairs on peak model performance is roughly as follows : Anthropic – Claude Fable 5 OpenAI – GPT 5.6 Sol Moonshot AI – Kimi K3 (open weights ) SpaceXAI – Grok 4.5 Zhipu (Z.ai) – GLM 5.2 (open weights) Meta – Muse Spark 1.1 DeepMind – Gemini Flash 3.5 Alibaba – Qwen 3.7 Max (3.8 announced, also to be open-weights, when writing) It is astonishing to see DeepMind, and some of the other American giants this low. In many ways, the X AI team deserves more credit. A visu

匹配字段正文

d AI | RAND; Frontier lab safety policies/commitments Responsible Scaling Policy | Anthropic; OpenAI — Preparedness Framework v2; Google DeepMind — Frontier Safety Framework v3.0; xAI — Risk Management Framework (RMF); Meta — Frontier AI Framework; Lab governance scrunity Anthropic is Quietly Backpedalling on its Safety Commitments — LessWrong; Holden Karnofsky on dozens of amazing opportunities to make AI safer — and all his AGI takes | 80,000 Hours (Can we trust Anthropic, or any AI company? 0

匹配字段正文

d GH200 GPUs, plus 1,328 installing. That is more than 11x the tracked A100 total of 41,208. The A100 chart now reads like a legacy fleet chart; Hopper is the live installed base. xAI Colossus 1 is the largest tracked Hopper deployment at 200,000 GPUs, and by itself is almost five times the entire tracked A100 set. Tesla Cortex follows at roughly 66,000 H100-equivalent GPUs, then Meta’s GenAI clusters at 49,152, a CoreWeave H200 cluster at 42,000, Voltage Park at 24,000, and Germany’s JUPITER Bo

匹配字段正文

llel reasoning, and on broader expertise in making generative models more efficient in training and inference across language and vision. What's next: Member of Technical Staff at xAI Yichen Xie Email: yichenxie0928@gmail.com Website: https://yichen928.github.io/ Advisor(s): Masayoshi Tomizuka Research Blurb: My research focuses on building multimodal foundation models and world models that understand and interact with complex physical environments. I aim to develop unified representations acros

匹配字段正文

being the creator of ChatGPT, and seems interested in AI as a truly general-purpose technology. Google’s thing is AI for science, and it’s often seen as having an edge on compute. xAI defies all expectations about when and where you can build data centers. Regardless of your opinions on these companies, their strategies and cultures can vary quite a bit, and it’s worth understanding the differences. These job postings tell us that the same is true for Chinese AI firms.7 Like Anthropic, Z.ai seem

匹配字段正文

pending an estimated $450bn a year on AI infrastructure. Its answer has been the EU AI Act and a capital commitment that is a rounding error by comparison. A single American site, xAI’s Colossus in Memphis, runs more than half a million GPUs. Europe has nothing remotely at that scale. The instinct to govern this technology is right, but we’re off on the ambition by orders of magnitude. It is fair to object that regulation is itself a form of power. But a rule book is not a substitute for the thi

匹配字段正文

This is catnip for engineers. It’s also especially critical for “hero run” large-scale training, so it draws a lot of investment and attention, e.g. recently as everyone dunked on xAI’s 10% MFU. But at the other end of the stack, there’s a more basic form of utilization that wrecks the relationship between achieved output and allocated capacity for inference workloads, GPU Allocation Utilization: GPU Allocation Utilization := GPU-seconds running application code ÷ GPU-seconds paid for Inference

匹配字段正文

and RadixArk co-hosted a GTC Happy Hour that brought together builders, researchers, and founders from across the inference and training ecosystem, including friends from OpenAI, xAI, DeepMind, Meta, NVIDIA, Ollama, and more. SGLang × RadixArk Happy Hour. The evening featured two technical spotlights: Banghua Zhu (RadixArk) introduced RadixArk and Miles, SGLang's native RL training framework purpose-built for large-scale MoE post-training workloads; Jason Zhao (ScitiX) presented SiMM, an open-s

匹配字段正文

a leading zero bug, and a decimal bug. However, it added an explanation of the problems it was trying to fix, aimed at a five-year-old, which made the issue quite clear. Also: Why xAI is giving you ‘limited’ free access to Grok 4 I still can’t decide if I think continuing the explain-to-a-five-year-old theme throughout the session was good conversational awareness, or overdone. For example, it correctly identified the Fibonacci sequence, and then went on to explain it at a five-year-old level. I

匹配字段正文

Our Work English; Español; 中文; Search Frontier AI developers such as OpenAI, Google, Anthropic, xAI, and others are governed by safety and security obligations under California’s SB 53, New York’s RAISE Act, Illinois’ SB 315, and the frontier AI part of the EU’s AI Act. These laws establish incident reporting requirements, model evaluation standards, safety and security mitigations, internal governance practices, and whistleblower protections. This document summarizes key provisions from these l

匹配字段正文

AM and bandwidth pressure, delivering rollout efficiency significantly higher than W8A8 (FP8 weights, FP8 activations) This project is jointly completed by the SGLang RL team, InfiXAI team, Ant Group Asystem & AQ Infra team, slime team, and RadixArk team. Related features and recipes have been synced to the slime and Miles communities. We welcome everyone to try them out and contribute. We are also further challenging ourselves with MXFP8 and NVFP4. We also gratefully acknowledge Verda Cloud for

匹配字段正文

密集型流程构建更可靠的自动化。 为何重要：这使 AI 从“问一个问题、得到一个答案”转向“描述你想要的结果，让模型围绕你现有的工具来编排工作”。 来源：VentureBeat – GPT‑5.2 来了：企业需要了解什么 AI 技术栈正在垂直整合，这改变了你的工具版图 摘要：近期一份对 AI 格局的分析认为，OpenAI、Anthropic、Google 和 xAI 正从“只做模型”转向垂直整合的技术栈，涵盖基础设施、模型、智能体和分发渠道，同时一个新的 Agentic AI Foundation 正试图防止碎片化。对高管而言，这意味着供应商选择不再仅仅关乎模型准确度，还关乎你希望自己的运营依托于哪个由智能体、连接器和合作伙伴应用构成的生态系统。 为何重要：你的 AI 战略正日益成为平台战略，因此围绕少数几个生态系统的整合将影响从采购到人才和集成选择的一切。 摘要：一篇 2025 年 12 月下旬的回顾指出，AI 系统已从静态内容生成器转变为能够使用工具、操作你的日历并跨软件协调以在无需持续提示的情况下完成任务的智能体。文章强调，人们日益期待在 2026 年，智能体将在后台越来越多地处理

匹配字段正文

not only refine AI models but also consider how to design structures within UX that prompt users to think again. AI & Society researchers[7] point out that simple explainability (XAI) or transparency alone is not enough to solve this problem. In fact, poorly designed explanations themselves can become another heuristic, reinforcing automation bias by fostering trust while reducing doubt. Accordingly, this post summarizes the five human–AI collaboration design implications presented in this revi

匹配字段正文