官方更新 DeepMind Blog Wed, 30 Sep 2026 正文 ok 208 words

Gemini 4 Argon:开启前沿智能的新时代

英文标题:Gemini 4 Argon: our next era of frontier intelligence

Gemini 4 Argon 凭借编程、推理、多模态能力以及处理长周期多步骤任务的能力,可胜任多种企业工作流。该模型在衡量真实世界长周期软件工程任务的 DeepSWE v1.1 上取得 77.9%,创下新纪录。在覆盖金融、编程、法律和税务工作的 Vals Index,以及 AutomationBench、LVBench 等评测中,Argon 均表现领先。其能力意味着开发者和企业可将其用于调试、代码库迁移、金融研究、法律研究与起草,以及基于视觉理解的知识工作。
打开原文相关 2.4 · 机会 1
media9.8 KB media2.0 KB media333.4 KB media40.1 KB media37.3 KB media38.9 KB media30.8 KB media35.8 KB media25.4 KB media45.8 KB media24.1 KB media43.4 KB media27.6 KB media49.4 KB media35.2 KB media24.6 KB media10.8 KB media28.4 KB media5.8 KB media12.7 KB

中文

赋能跨领域的编程与企业工作流

Gemini 4 Argon 在编程、推理和多模态方面的能力,以及持续完成长时间、多步骤任务的能力,使其能够在一系列企业工作流中表现出色。

Google 工程师已在日常工作中使用 Argon,覆盖从常规调试到大规模代码库迁移和算法设计等任务。它在 DeepSWE v1.1 上创下新的最佳成绩(77.9%);该评测用于衡量模型在真实世界、长周期软件工程任务中的表现。

除编程外,Argon 还是 Vals Index 上表现最领先的模型。该指数衡量模型在金融、编程、法律和税务工作中的经济影响,并根据各行业对美国 GDP 的贡献进行加权。我们也在其他特定领域评测中看到其同样领先的表现,例如 Vals Finance Agent v2(多步骤金融研究)以及 Harvey 的 Legal Agent Benchmark(法律研究和文书起草)。在 Zapier 用于衡量核心业务职能端到端执行能力的基准测试 AutomationBench 中,Argon 以 51.3% 的得分位居第一。

当知识工作需要视觉理解能力时,Argon 同样具备独特优势。它能够完成专业图表分析、从长视频中识别细节,并根据一系列文档采取行动。例如,在衡量长视频理解能力的 LVBench 上,Argon 以 91.7% 的得分达到当前最佳水平。

English / 原文

Enabling coding and enterprise workflows across domains

Gemini 4 Argon’s capabilities across coding, reasoning, and multimodality and its ability to sustain long, multi-step tasks enable it to excel across a range of enterprise workflows.

Google engineers have been using Argon for their daily tasks, from everyday debugging to large-scale codebase migrations and algorithm designs. It sets a new state of the art on DeepSWE v1.1 (77.9%), which measures a model’s performance in real-world long-horizon software engineering tasks.

Beyond coding, Argon is the leading model on the Vals Index, which measures economic impact across finance, coding, legal, and tax work, with every sector weighted by its contribution to U.S. GDP. We see similarly leading performance across other domain specific evaluations, like Vals Finance Agent v2 (multi-step financial research) and Harvey’s Legal Agent Benchmark (legal research and drafting). On AutomationBench, Zapier’s benchmark measuring end-to-end execution across core business functions, Argon ranks #1 with a score of 51.3%.

Argon is also uniquely strong when knowledge work requires visual understanding. It’s able to drive professional chart analysis, identify details from long videos, and take action based on a series of documents. For example, on LVBench, which measures long video understanding, Argon is state of the art with a score of 91.7%.