DeepSeek最新动作。 12月1日晚间， DeepSeek 发布两个正式版模型： DeepSeek-V3.2 和 DeepSeek-V3.2-Speciale 。 据悉，目前DeepSeek官方网页端、App 和 API 均已更新为正式版 DeepSeek-V3.2。Speciale 版本目前仅以临时 API 服务形式开放，以供社区评测与研究。 DeepSeek-V3.2-Speciale 的目标是将开源模型的推理能力推向极致，探索模型能力的边界，是 DeepSeek-V3.2 的长思考增强版，同时结合了 DeepSeek-Math-V2 的定 理证明能力。 据DeepSeek透露，DeepSeek-V3.2-Speciale 在主流推理基准测试上的性能表现媲美 Gemini-3.0-Pro。同时，V3.2-Speciale 模型成功斩获 IMO 2025（国际数学奥林匹 克）、CMO 2025（中国数学奥林匹克）、ICPC World Finals 2025（国际大学生程序设计竞赛全球总决赛）及 IOI 2025（国际信息学奥林匹克）金牌。其中，ICPC 与 IOI 成绩分别达到了人类选手第二名与第十名 ...

2025.12. 01 两款模型有着不同的定位。DeepSeek-V3.2的目标是平衡推理能力与输出长度，适合日常使用，例 如问答场景和通用智能体任务场景。9月底DeepSeek发布了实验版V3.2-Exp，此次是正式版更 新。在公开推理测试中，V3.2达到了GPT-5的水平，仅略低于谷歌的Gemini3 Pro。 本文字数：1580，阅读时长大约3分钟 作者 | 第一财经 刘晓洁 12月1日晚，DeepSeek又上新了两款新模型，DeepSeek-V3.2和DeepSeek-V3.2-Speciale， 在推理能力上全球领先。 据DeepSeek公布的数据，Speciale在多个推理基准测试中超越谷歌最先进的Gemini3 Pro。具体 来看，在美国数学邀请赛、哈佛MIT数学竞赛、国际奥林匹克数学竞赛等测试中，V3.2-Speciale都 超过了Gemini3 Pro，但在编程、理工科博士生测试中略逊于谷歌。 DeepSeek-V3.2-Speciale则是此次的重头戏，其目标是"将开源模型的推理能力推向极致，探索 模型能力的边界"。据介绍，Speciale是V3.2的长思考增强版，同时结合了DeepSee ...

Core Insights - DeepSeek has officially released two models: DeepSeek-V3.2 and DeepSeek-V3.2-Speciale, with updates available on the official website, app, and API [1] - DeepSeek-V3.2 aims to balance reasoning capabilities and output length, making it suitable for everyday use cases such as Q&A and general agent tasks [1] - DeepSeek-V3.2-Speciale is designed to push the reasoning capabilities of open-source models to the limit, enhancing long-thinking abilities and incorporating theorem-proving capabilities from DeepSeek-Math-V2 [1] Model Performance - The V3.2-Speciale model exhibits excellent instruction-following, rigorous mathematical proof, and logical verification capabilities, performing comparably to leading international models on mainstream reasoning benchmarks [1] - Notably, the V3.2-Speciale model has achieved gold medals in several prestigious competitions, including IMO 2025, CMO 2025, ICPC World Finals 2025, and IOI 2025 [1] - In the ICPC and IOI competitions, the model's performance reached the level of the second and tenth place among human competitors, respectively [1]

Core Insights - DeepSeek has launched two new models: DeepSeek-V3.2 and DeepSeek-V3.2-Speciale, aiming to enhance reasoning capabilities and output length for various applications [1][2][3] Model Features - DeepSeek-V3.2 is designed for everyday use, balancing reasoning ability and output length, and has achieved performance comparable to GPT-5 in benchmark tests [2][3] - DeepSeek-V3.2-Speciale enhances long reasoning capabilities and incorporates theorem proving abilities from DeepSeek-Math-V2, excelling in complex tasks but requiring more tokens and higher costs [3][4] Technological Advancements - DeepSeek-V3.2 is the first model to integrate reasoning with tool usage, supporting both reasoning and non-reasoning modes for tool invocation, significantly improving generalization capabilities [4] - The model has been trained on over 1,800 environments and 85,000 complex instructions, narrowing the performance gap between open-source and closed-source models [4] Market Outlook - The AI industry is experiencing a resonance period, with rapid expansion in AI infrastructure and commercialization of downstream applications, expected to continue thriving through 2026 [5][6][7] - Investment opportunities are identified in domestic AI chains, overseas AI hardware markets, and innovative applications in the domestic edge AI industry [7]

Core Insights - DeepSeek has launched two new models, DeepSeek-V3.2 and DeepSeek-V3.2-Speciale, which are leading in reasoning capabilities globally [1][3]. Model Overview - DeepSeek-V3.2 aims to balance reasoning ability and output length, suitable for everyday use such as Q&A and general intelligence tasks. It has reached the level of GPT-5 in public reasoning tests, slightly below Google's Gemini3 Pro [3]. - DeepSeek-V3.2-Speciale is designed to push the reasoning capabilities of open-source models to the extreme, combining features from DeepSeek-Math-V2 for theorem proving, and excels in instruction following and logical verification [3][4]. Performance Metrics - Speciale has surpassed Google's Gemini3 Pro in several reasoning benchmark tests, including the American Mathematics Invitational, Harvard MIT Mathematics Competition, and International Mathematical Olympiad [4]. - In various benchmarks, DeepSeek's performance is competitive, with specific scores noted in a comparative table against GPT-5 and Gemini-3.0 [5]. Technical Limitations - Despite achievements, DeepSeek acknowledges limitations compared to proprietary models like Gemini3 Pro, particularly in knowledge breadth and token efficiency [6]. - The company plans to enhance pre-training computation and optimize reasoning chains to improve model efficiency and capabilities [6][7]. Mechanism Innovations - DeepSeek introduced a Sparse Attention Mechanism (DSA) to reduce computational complexity, which has proven effective in enhancing performance without sacrificing long-context capabilities [7][8]. - Both new models incorporate this mechanism, making DeepSeek-V3.2 a cost-effective alternative that narrows the performance gap with proprietary models [8]. Community Reception - The release has been positively received in the community, with users noting that DeepSeek's models are now comparable to GPT-5 and Gemini3 Pro, marking a significant achievement in open-source model development [8].

Core Insights - DeepSeek has released the official version of its V3.2 model, which significantly enhances reasoning and agent capabilities compared to previous versions [2][9] - The V3.2-Speciale version is an open-source model that performs comparably to Gemini-3.0-Pro on mainstream reasoning benchmarks and has achieved gold medal levels in several prestigious competitions [3][11] - The integration of the DeepSeek Sparse Attention (DSA) technology in V3.2 improves long text processing efficiency and reduces costs by over 50% [3][10] Model Development - The V3 series has been iterated over the past year, with V3.2 being the latest release, focusing on unifying thinking and non-thinking models, a trend seen in other closed-source models like Gemini and GPT-5 [6][9] - The release timeline for DeepSeek models in 2025 includes various versions, each with specific enhancements, such as the introduction of DSA in V3.2 for stability and reasoning improvements [7][8] Performance Metrics - DeepSeek-V3.2 has achieved reasoning capabilities on par with GPT-5 and has shown significant improvements in output length and computational efficiency compared to Kimi-K2-Thinking [10][14] - The V3.2-Speciale version excels in complex tasks, achieving high scores in various academic competitions, including IMO 2025 and ICPC 2025, with notable rankings among human competitors [11][14] Tool Utilization - A key advancement in V3.2 is the incorporation of thinking processes into tool calls, allowing the model to support both thinking and non-thinking modes in its operations [15][18] - DeepSeek has developed a large-scale agent training data synthesis method that enhances the model's generalization capabilities by creating numerous "hard-to-answer, easy-to-verify" tasks [16][18]

Core Viewpoint - DeepSeek has launched two new models, DeepSeek-V3.2 and DeepSeek-V3.2-Speciale, aiming to enhance reasoning capabilities and application in various scenarios [1][2]. Model Features - DeepSeek-V3.2 aims to balance reasoning ability and output length, suitable for daily use such as Q&A and general agent tasks. It has achieved performance comparable to GPT-5 in benchmark tests, slightly below Gemini-3.0-Pro [2]. - DeepSeek-V3.2-Speciale is an enhanced version of V3.2, integrating theorem proving capabilities from DeepSeek-Math-V2, excelling in instruction following and logical verification. It has won gold medals in several prestigious competitions, including IMO 2025 and ICPC World Finals 2025 [3]. Technological Advancements - DeepSeek-V3.2 is the first model to integrate reasoning with tool usage, supporting both reasoning and non-reasoning modes for tool invocation. The model has been trained on over 85,000 complex instructions across 1,800 environments, significantly improving its generalization ability [4]. Market Outlook - The AI industry is entering a period of resonance, with rapid expansion in AI infrastructure and commercialization of downstream applications. Analysts predict continued prosperity in the AI sector, with a focus on domestic chips, servers, and AI applications [5][7]. - On December 1, AI-related stocks showed strong performance in the secondary market, with significant gains in AI mobile devices and AI computing power sectors [6].

①DeepSeek V3.2正式版发布：强化Agent能力，融入思考推理； DeepSeek V3.2正式版发布；豆包手机助手发布技术预览版。 ■ 汇丰银行与Mistral Al达成战略合作 日本报 li l = ray 滴滴自动驾驶在广州试运行 全天候、全无人Robotaxi服务 ②豆包手机助手发布技术预览版； ③清华大学成立具身智能与机器人研究院； ④滴滴自动驾驶在广州试运行全天候、全无人Robotaxi服务； ⑤汇丰银行与Mistral AI达成战略合作，以增进银行业务中生成式AI的应用。 ■ 来源:一财快讯 ...

衡宇 发自 奥特赛德 量子位 | 公众号 QbitAI 突袭！ ChatGPT发布三周年，DeepSeek嚯一下发出两个模型： 前者聚焦平衡实用 ，适用于日常问答、通用Agent任务、真实应用场景下的工具调用。 推理达GPT-5水平，略低于Gemini-3.0-Pro。 下图展示的是DeepSeek-V3.2与其他模型在各类Agent工具调用评测集上的得分 ——特别强调，DeepSeek-V3.2并没有针对这些测试集的工具做特殊训练。 划重点，ICPC达到人类选手第二、IOI人类选手第十名水平。 具体来说，DeepSeek-V3.2侧重于平衡推理能力与输出长度，降低计算开销。 DeepSeek官微推文中写道，"DeepSeek-V3.2模型在Agent评测中达到了当前开源模型的最高水平"。 该模型其他情况如下： DeepSeek-V3.2 DeepSeek-V3.2-Speciale 推理能力比肩GPT-5； 相比Kimi-K2-Thinking大幅缩短输出长度，减少用户等待时间； DeepSeek旗下首个"思考融入工具调用" 的模型，支持思考/非思考双模式工具调用； 基于1800+环境、85000+复杂指令 ...

Core Insights - DeepSeek has released two official model versions: DeepSeek-V3.2 and DeepSeek-V3.2-Speciale [1][4] Group 1: Model Features - DeepSeek-V3.2 aims to balance reasoning capability and output length, making it suitable for everyday use. In benchmark tests, it reached the level of GPT-5, slightly below Gemini-3.0-Pro, while significantly reducing output length compared to Kimi-K2-Thinking, thus decreasing computational overhead and user wait time [3] - DeepSeek-V3.2-Speciale is designed to push the reasoning capabilities of open-source models to the extreme, exploring the boundaries of model capabilities. It is an enhanced version of DeepSeek-V3.2 with long-thinking capabilities and incorporates theorem-proving abilities from DeepSeek-Math-V2 [3] Group 2: Achievements - The V3.2-Speciale model has successfully won gold medals at several prestigious competitions, including IMO 2025, CMO 2025, ICPC World Finals 2025, and IOI 2025 [3] Group 3: Availability - The official web version, app, and API have all been updated to the official version of DeepSeek-V3.2. The Speciale version is currently available only as a temporary API service for community evaluation and research [4]