搜索“OpenAI”

o ensure accuracy, clarity, and responsible discussion of advanced AI concepts. Discover More Square Launches Flagship Kiosk With Agentic AI Voice Ordering; NVIDIA Unveils cuPotential as Semiconductor Leaders Adopt CUDA-X; Microsoft Research Publishes AI Weather Forecasting Study in Nature; Google Rolls Out AI Scam Defenses Across Android Messages and Calls; OpenAI Alleges Employee Who Rejoined Apple Took Unreleased Device Files; Baseten Adds Runway’s Gen-4.5 and Aleph 2.0 to Dedicated Inference

匹配字段正文

Muse Spark是团队推出的首个模型系列，此次推出的Muse Wonder系列规模更大、能力更先进，引领着Meta的模型能力迈入前沿水平。 2025年6月，扎克伯格重金组建超级智能团队，投百亿美元并购AI创企、狂揽顶尖AI人才，今日终于放出“大招”。今年4月份亮相的Muse Spark主打低延迟、个人智能助手任务，今日亮相的Muse Wonder则正式向OpenAI、Anthropic的顶级模型发起挑战。 据Meta介绍，Muse Wonder系列引入主动记忆（Active Memory）和主动专注（Active Focus）能力，能够持续跟踪用户关注的事项，并围绕用户目标推进任务。Meta称，这也是其迈向“个人超级智能”（Personal Superintelligence）的重要一步。 在Meta公布的长流程编程测试DeepSWE 1.1和SWE-Marathon超长流程软件工程测试中，Muse Wonder Full得分超过了Claude Fable 5.1、GPT-6 Astra以及Claude Opus 5.5，展现出较强的Agent任务执行能力。 在Artificial

匹配字段正文

on feature (rewriting instructions via an LLM for better editing results), set up one of the supported LLM providers: Option 1: Gemini (default) Option 2: MiniMax Uses the MiniMax OpenAI-compatible API with the MiniMax-M2.7 model by default. You can also use MiniMax-M2.7-highspeed for faster responses. Option 3: Any OpenAI-compatible API Note: The LLM API is not required for basic usage. Without it, the Agent will still perform ROI detection and image stitching normally, but will skip the instru

匹配字段正文

ry and OpenRouter, according to the company. Microsoft built the model by post-training Qwen3.5-9B and said it plans to rebase it on other models, including Microsoft AI (MAI) and OpenAI models. Model targets specific decisions Decision-1 supports yes-or-no judgments, choices among predefined options, ratings, and rubric-based assessments of AI responses and agent actions, Microsoft said. Developers can use its outputs to classify inputs, rank work, verify responses, or determine the next action

匹配字段正文

liar tasks that weren't part of training optimization, and credits this to a stronger base model. On the Vals AI Index , Muse Wonder Full (Max) debuts at fifth place and beats all OpenAI models, according to Vals AI. Only current Anthropic models score higher. Muse Wonder lands at tenth. The biggest improvements come from coding and math. On ProofBench, a private benchmark for mathematical proofs, Muse Wonder Full solves 94 out of 100 problems, up from 58 for its predecessor. That puts it just f

匹配字段正文

lete. We have been coveringe MSL’s 14 month journey to becoming a frontier lab, but only today did the final ball drop: Meta’s Muse Wonder to match Spark and Glimmer, the way that OpenAI has Sol/Terra/Luna and Anthropic has Opus/Sonnet/Haiku. Interestingly, while Muse doesn’t claim a specific size, Muse Wonder Full does beat Fable and Astra on some benchmarks, and with various degrees of posttraining it is a little unclear if Wonder is more of a Fable/Astra class model or not. We have included a

匹配字段正文

Kopparapu in a video accompanying the launch. The new models are the first frontier model release since Scale AI founder Alexandr Wang restructured Meta’s AI efforts in August of last year . Since then, Meta has released its lightweight Muse Spark model and Muse agent but has avoided the kind of state-of-the-art model releases OpenAI and Anthropic are known for. According to the post, the models were trained at Meta’s Prometheus data center in central Ohio , which came online earlier this year.

匹配字段正文

on one’s own hardware, was widely adopted by researchers and startups. But the company fell behind after delivering several disappointing versions of the model, and as Anthropic, OpenAI, and Google rapidly introduced more capable models. Meta announced the first model from its new lab, Muse Spark, in April of this year. The company has also released a small open-weight model called Muse Glimmer, and says it plans to release larger open-weight models in the future. It did not disclose details of

匹配字段正文

源模型系列，包括Muse Wonder和Muse Wonder Full。Meta表示，这两款模型是公司旗下目前能力最强的模型，在个人与专业智能体任务中均达到业界领先水平。Muse Wonder在公司内部代号为“西瓜”，相较此前的模型实现了显著进步，具备“前沿能力”。 Meta公司管理层认为，Muse Wonder在编程以及处理金融、法律问题等方面，已能够与OpenAI的ChatGPT模型和Anthropic的Claude模型竞争。 Meta表示，这两款模型标志着公司在模型规模化发展道路上迈出了新的一步，也代表着公司向“个人超级智能”迈出的重要一步。Meta AI负责人Alexandr Wang表示：“这些是真正的前沿模型。我们确实认为，它们能够与市面上最优秀的模型竞争，尤其是在个人与专业智能体任务方面，达到了最先进的水平。” 扎克伯格在接受外媒采访时指出，Muse Wonder“将使AI展现出更加复杂的行为能力，达到当前AI领域所能实现的前沿水平”：“要达到目前这一水平，既需要大量算力和资金投入，也需要研发团队在较长一段时间内开展严谨、扎实的科学研究。” Meta从即日起通过Meta

匹配字段正文

V2 pipeline verification is also available; see python inference/inference hf.py --help. SGLang serving inference/sglang client.py is a single-file, standard-library client for an OpenAI-compatible endpoint. Use a dedicated serving environment and follow the minimal server recipe: Then replay the packaged mobile, desktop, and web requests: Each JSON embeds its screenshot and an evaluated current-image-only multi-turn request. The packaged step-$t 0$ cases use the exact role sequence system, user

匹配字段正文

mework unless their individual requirements state otherwise. Installation Python 3.10 or newer is required. All commands below are executed from the repository root. Configure the OpenAI-compatible model service through MODEL URL, MODEL NAME, and either API KEY or MODEL API KEY; the same values can also be edited at the top of each domain script. API KEY takes precedence when both key variables are set. Set MODEL NAME to the served name of your UI-Venus-2 9B or 27B model. Run multi-turn inferenc

匹配字段正文

read We have been coveringe MSL’s 14 month journey to becoming a frontier lab, but only today did the final ball drop: Meta’s Muse Wonder to match Spark and Glimmer, the way that OpenAI has Sol/Terra/Luna and Anthropic has Opus/Sonnet/Haiku. Interestingly, while Muse doesn’t claim a specific size, Muse Wonder Full does beat Fable and Astra on some benchmarks, and with various degrees of posttraining it is a little unclear if Wonder is more of a Fable/Astra class model or not. We have included a

匹配字段正文

model sales while strengthening its proprietary AI productivity tools. Meta is seeking to expand its monetization opportunities in the enterprise AI market, where it competes with OpenAI, Anthropic PBC, and Google LLC. The company has until now focused on expanding its ecosystem through Llama, its open AI model family. The standout features of the Muse Wonder series are Active Memory and Active Focus. Active Memory proactively retains information from previous conversations and important user co

匹配字段正文

实测翻车，官方承认短板 对比视频显示Wonder Full生成的3D割草机游戏在视觉细节上远逊于Opus 5.5，被调侃为AAA与Roblox的差距；Meta模型卡承认3D任务是已知局限，Alexandr Wang回应称团队未针对此类任务专门训练。2026-10-13 10:10小扎打赢AI翻身仗？空降新模型长程编程夺冠，3D实测却惨遭打脸AI前线速览 在被OpenAI、Anthropic轮番甩开之后，扎克伯格终于有底气宣布：Meta重新回到了AI技术前沿。 北京时间10月13日凌晨，Meta Superintelligence Labs（MSL）正式推出新一代旗舰模型系列Muse Wonder和Muse Wonder Full，直接向GPT-6 Astra、Claude Opus 5.5等顶级模型发起挑战。 第三方评测机构Artificial Analysis同步公布的数据显示，Muse Wonder Full在综合智能指数中拿下56分，距离目前58分的Claude Opus 5.5仅差2分，并与Claude Sonnet 5.5处于同一分数档位。在多项长程编程、Agent任务测试中

匹配字段正文

okens and reasoning effort. Step 6. Copy working code. Click Get code for a curl or Python snippet. It calls the Responses endpoint at https://api.meta.ai/v1/responses. The API is OpenAI-compatible, so the OpenAI SDK works with a new base URL: Step 7: Or use Muse Code in your terminal. On Mac, install with curl -fsSL https://dev.meta.ai/install.sh | bash. On Windows, use irm https://dev.meta.ai/install.ps1 | iex. Run muse in your project folder, trust the workspace, and sign in. The default mode

匹配字段正文

财联社10月14日（编辑 史正丞）紧挨着周一发布前沿模型的Meta， 科技巨头谷歌周二宣布，公司备受期待的旗舰大模型Gemini 4 Argon正式上架。 （来源：谷歌掌门皮查伊） 这也意味着，现在能提供“比肩”Anthropic和OpenAI旗舰大模型服务的公司，在本周头两天里就增加了两家，再一次体现出AI领域的实力格局能以多快的速度发生变化。 作为背景，Argon是谷歌去年推出Gemini 3以来的首个重大前沿模型。 此前谷歌披露的性能数据显示，与主要竞品对比，Argon更擅长金融、法律、长资料分析等专业知识工作。同时，谷歌把单次输出上限从6.4万提高到100万词元（注：GPT-6 Astra和Claude Opus 5.5常规输出上限均为12.8万词元），给长时间推理、多轮验证和大型代码修改留出更多空间。 与此同时，Argon首发优惠价只有Astra和Fable的五分之一，即百万词元的输入/输出价格仅为2/10美元，11月底优惠结束后的常规定价（4/20美元）则与Opus 5.5持平。 目前Argon模型只向每月至少付费20美元的订阅用户以及API用户开放。谷歌宣布，该模型周二起

匹配字段正文

ory $2 per million input tokens and $10 per million output tokens through November 30, with a single 1 million-token budget for input and output combined. That’s the same price as OpenAI’s GPT-6.1 Sol. After that date, Google’s new model will cost $4 and $20 per million tokens, matching Anthropic’s Claude Opus 5.5 . Cached input tokens are 95 percent off during the introductory period. During the introductory period, Argon also costs a fifth of what Anthropic charges for Claude Fable 5.1 and Ope

匹配字段正文

.” Google’s chart at the Sept. 30 unveiling had the model at 77.9% on DeepSWE v1.1, a test of long-horizon software engineering, against 74.2% for Anthropic PBC’s Claude Opus 5.5. OpenAI Group PBC’s GPT-6 Astra sat a 10th of a point further back. The agentic coding interface from Google Antigravity now runs directly inside the Gemini app for macOS, which means a chat can turn into a coding session without a jump to a separate tool. Google did not mention a Windows version. The app was one part o

匹配字段正文

by AI agents making unwanted incursions — Big Tech leaders have pivoted to speak about AI’s ability to cure cancer, for example. Gemini 4 Argon competes with frontier models like OpenAI’s GPT-6 Astra and Meta’s new Muse Wonder. Google had the edge among AI companies for most of 2025, but advancements from other labs left it mostly behind for this year. Also throughout the past year, concerns about AI safety have risen significantly, particularly around “AI doomsday” scenarios. Google says that

匹配字段正文

nstall the latest version of transformers (compatible with transformers =5.0.0 ): Then load and run the model: Use it in your agent harness Serve LFM2.5-24B-A2B locally behind any OpenAI-compatible backend, then point your agent harness at it. For example, with Hermes Agent: The port depends on your serving backend: llama.cpp and MLX use 8080 , vLLM uses 8000 , SGLang uses 30000 , and LM Studio uses 1234 . See the model card for OpenClaw and Pi configs, and our guide for the full setup. Check ou

匹配字段正文