男生普通话版 下载mp3 女声普通话版 下载mp3 粤语版 下载mp3 西南方言版 下载mp3 东北话版 下载mp3 上海话版 下载mp3 DeepSeek-V3.1正式发布 市场盘点 今日优选 美国司法部官员敦促鲍威尔解雇理事库克 美联储官员依旧对降息保持谨慎，下任美联储主席候选人则呼吁大幅降息 美国8月标普全球综合PMI创8个月新高，交易员减少对美联储今年两次降息的押注 以总理内塔尼亚胡：要在以方可接受条件下停战 美欧正式敲定贸易协议框架 特朗普政府考虑使用《芯片法案》中的20亿美元资金用于购买芯片相关的关键矿产 5000亿"准财政"工具将出炉 中国月度用电量首破万亿大关 周四，美元指数持续走强，盘中一度升至98.67的日内高点，最终收涨0.43%，报98.62；基准的10年期美债收益率收报4.330%，2年期美债收益率收报 3.794%。 现货黄金开盘跳涨后持续下跌，盘中一度跌至3325.19美元的日内低点，随后收复部分失地，最终收跌0.29%，收报3338.55美元/盎司；现货白银收涨0.56%， 报38.10美元/盎司。 由于俄罗斯和乌克兰相互指责对方导致和平进程停滞不前，而且早些时候的美国数据显示 ...

Core Insights - DeepSeek officially released DeepSeek-V3.1 on August 21, 2023 [1] - The upgrade includes significant enhancements in search evaluation metrics, particularly in complex search tests and multidisciplinary expert-level challenges [3][4] Upgrade Features - The official app and web model have been upgraded to DeepSeek-V3.1, allowing users to switch between thinking and non-thinking modes via a "Deep Thinking" button [2][3] - DeepSeek API has also been upgraded, with deepseek-chat corresponding to non-thinking mode and deepseek-reasoner corresponding to thinking mode, expanding context to 128K [3] - The API Beta interface supports strict mode Function Calling to ensure outputs meet schema definitions [3] Performance Improvements - DeepSeek-V3.1 shows significant performance improvements over R1-0528 in complex search tests (browsecomp) and multidisciplinary expert-level problem tests (HLE) [3][4] - The model utilizes UE8M0 FP8 Scale parameter precision, with substantial adjustments made to the tokenizer and chat template, resulting in noticeable differences from DeepSeek-V3 [3] Market Reaction - DeepSeek concept stocks experienced a sharp rise in the market towards the end of the trading day [3]

Market Performance - The main board led the gains with notable increases in stocks such as Yuyin Co. (+10.05%), Tianrongxin (+10.01%), Zhuoyue Technology (+10.00%), Jishi Media (+10.00%), and Kesen Technology (+9.99%) [1] - The ChiNext board saw significant growth with stocks like Beixin Source (+19.97%), Chuangyitong (+15.01%), and Meiri Interactive (+13.62%) [1] - The Sci-Tech Innovation board also performed well, with Sanwei Xinan (+20.01%), Xin'an Century (+11.00%), and Meidike (+7.25%) leading the way [1] - Active sub-industries included SW Games III (+1.64%) and SW General Publishing (+1.58%) [1] Domestic News - Shanghai Xiangyao Display Technology Co., Ltd. successfully completed a financing round of several hundred million RMB, aimed at developing MicroLED micro-display technology and expanding production capacity to meet the growing demand in the global AR+AI terminal market [1] - Chip equipment manufacturer Chip Micro announced significant market breakthroughs with its wafer-level and board-level direct-write lithography equipment, securing purchase orders from several leading domestic packaging and testing companies [1] - Vivo's Vision Exploration Edition was unveiled, featuring a Micro-OLED dual 8K display with 94% DCI-P3 color gamut coverage and professional-grade color accuracy [1] - A report from KISTEP indicated that China has surpassed South Korea in semiconductor storage technology, although Samsung and SK Hynix still maintain a leading position globally [1] Company Announcements - Century Dingli reported a total operating income of 128 million RMB for H1 2025, a decrease of 6.87% year-on-year, but a net profit increase of 160.47% to 9 million RMB [2] - Lante Optics achieved an operating income of 577 million RMB in H1 2025, a year-on-year increase of 52.54%, with a net profit of 103 million RMB, up 110.27% [2] - Xintong Electronics reported an operating income of 448 million RMB for H1 2025, a 10.85% increase year-on-year, with a net profit of 50 million RMB, up 5.83% [2] - Boch Electronic's H1 2025 operating income reached 1.103 billion RMB, a 24.89% year-on-year increase, with a net profit of 640 million RMB, up 30.32% [2] Overseas News - TrendForce research indicates that with the release of NVIDIA's GB200 NVL72 server in 2025, the penetration rate of liquid cooling technology in AI data centers is expected to rise from 14% in 2024 to 33% in 2025 [2] - According to South Korean customs data, the export value of South Korean memory chips reached 21.329 billion USD in Q2 2025, a year-on-year increase of 20.7% and a quarter-on-quarter increase of 31.8% [2] - AMD has announced the discontinuation of the B650 chipset, with motherboard manufacturers no longer able to procure it, leading to a phase of inventory clearance [2] - Google has launched the Pixel 10 series in the U.S. market, eliminating physical SIM card slots in favor of eSIM technology, allowing for the simultaneous use of two eSIMs [2]

人工智能迎来两则重磅利好。 首先是，政策层面的利好。8月21日，安徽省人民政府印发《打造通用人工智能产业创新和应用高地若干政策（2.0版）》 ，其中提出了9条具体政策举措，加速大 模型等人工智能技术赋能千行百业，推动安徽省率先进入通用人工智能时代。 在产业层面，DeepSeek今日宣布，DeepSeek-V3.1正式发布。 受此消息影响，DeepSeek概念股——每日互动尾盘直线拉升，截至收盘，大涨13.62%。DeepSeek在文 章中提到，DeepSeek-V3.1使用了UE8M0 FP8 Scale的参数精度。DeepSeek官微在置顶留言里表示，UE8M0 FP8是针对即将发布的下一代国产芯片设计。 安徽重磅发布 8月21日，安徽省人民政府印发《打造通用人工智能产业创新和应用高地若干政策（2.0版）》（以下简称《若干政策》），其中提到，充分发挥财政资金等引导作 用，撬动保险、信贷、基金等社会资本投向通用人工智能产业。加快运营总规模不低于200亿元的省人工智能产业主题基金，以参股方式支持市国资平台、经营主 体等设立通用人工智能领域子基金，满足企业和项目资金需求。建立健全保护改革、鼓励探索、宽容失误、纠正 ...

Core Insights - DeepSeek officially launched DeepSeek-V3.1 on August 21, featuring significant upgrades in its architecture and performance [1]. Group 1: Major Changes in DeepSeek-V3.1 - The new hybrid reasoning architecture allows the model to support both thinking and non-thinking modes simultaneously [2]. - Enhanced thinking efficiency enables DeepSeek-V3.1-Think to provide answers in a shorter time compared to DeepSeek-R1-0528 [2]. - Improved agent capabilities through post-training optimization have led to better performance in tool usage and agent tasks [2]. Group 2: API and User Experience Enhancements - The official app and web model have been upgraded to DeepSeek-V3.1, allowing users to switch freely between thinking and non-thinking modes via a "deep thinking" button [2]. - The DeepSeek API has also been upgraded, with deepseek-chat corresponding to non-thinking mode and deepseek-reasoner to thinking mode, expanding context to 128K [2]. - The API Beta interface now supports strict mode function calling to ensure outputs meet schema definitions [2]. Group 3: Performance Metrics - DeepSeek-V3.1 has shown significant improvements in multiple search evaluation metrics, outperforming R1-0528 in complex search tests requiring multi-step reasoning and expert-level multidisciplinary challenges [2]. Group 4: Technical Adjustments - DeepSeek-V3.1 utilizes UE8M0FP8Scale parameter precision and has made substantial adjustments to the tokenizer and chat template, resulting in noticeable differences from DeepSeek-V3 [3]. Group 5: Market Reaction - Following the announcement of DeepSeek-V3.1, DeepSeek concept stock Daily Interaction (300766) experienced a sharp rise in the late trading session [3].

Core Insights - DeepSeek officially announced the release of version 3.1 on August 21, featuring significant upgrades including a hybrid reasoning architecture and improved response efficiency [1] - The pricing for API calls has increased, with new rates effective from September 6, including a removal of night-time discounts [2] - The market reacted positively to the news, with shares of Daily Interactive (300766.SZ) rising by 13.62% to 47.98 CNY per share [4] Group 1: Product Updates - DeepSeek V3.1 introduces a hybrid reasoning architecture that allows for flexible switching between thinking and non-thinking modes, enhancing agent capabilities [1] - The model utilizes UE8M0 FP8 Scale parameter precision and has undergone significant adjustments to the tokenizer and chat template, marking a clear distinction from DeepSeek V3 [1] - The foundational model of V3.1 has been retrained with an additional 840 billion tokens, and both the foundational and post-training models are available on Huggingface and MoDa [4] Group 2: Pricing Changes - The API call pricing has been adjusted, with input prices set at 0.5 CNY per million tokens for cache hits and 4 CNY per million tokens for cache misses, up from 2 CNY for V3 [2] - The output price has increased to 12 CNY per million tokens, compared to 8 CNY for V3 [2] Group 3: Market Reaction - Following the announcement, shares of Daily Interactive surged, attributed to its perceived connection with DeepSeek, as it was rumored to hold a 14.50% stake in the DeepSeek development team [4] - Daily Interactive clarified that it does not hold any equity in DeepSeek or its associated companies, despite previous speculation [4]

Core Viewpoint - DeepSeek officially released DeepSeek-V3.1 on August 21, showcasing significant upgrades in its capabilities and performance metrics [1]. Summary by Sections Upgrade Features - The official app and web model have been upgraded to DeepSeek-V3.1, allowing users to switch between thinking and non-thinking modes via a "Deep Thinking" button [2]. - The DeepSeek API has also been upgraded, with deepseek-chat corresponding to non-thinking mode and deepseek-reasoner to thinking mode, expanding context to 128K [2]. - The API Beta interface supports strict mode Function Calling to ensure outputs meet schema definitions [2]. Performance Improvements - DeepSeek-V3.1 has shown substantial improvements in multiple search evaluation metrics, particularly in complex search tests requiring multi-step reasoning (browsecomp) and expert-level multidisciplinary problem tests (HLE), outperforming R1-0528 significantly [2]. - The model utilizes UE8M0 FP8 Scale for parameter precision, and there have been major adjustments to the tokenizer and chat template, resulting in noticeable differences from DeepSeek-V3 [2]. Market Reaction - DeepSeek concept stocks experienced a sharp rise in the market towards the end of the trading session [3]. Architectural Enhancements - The new mixed reasoning architecture supports both thinking and non-thinking modes within a single model [5]. - Enhanced thinking efficiency allows DeepSeek-V3.1-Think to provide answers in a shorter time compared to DeepSeek-R1-0528 [5]. - The agent capabilities have improved significantly through post-training optimization, enhancing performance in tool usage and intelligent agent tasks [5].

Core Viewpoint - DeepSeek officially announced the release of version V3.1, which includes significant upgrades such as a hybrid reasoning architecture and enhanced agent capabilities, leading to improved response efficiency [1][4]. Pricing Adjustments - DeepSeek has increased the API interface call prices effective from September 6. The new pricing structure includes a charge of 0.5 yuan per million tokens for cache hits and 4 yuan per million tokens for cache misses, up from 2 yuan in V3. The output price has also risen to 12 yuan per million tokens from 8 yuan in V3 [2]. Market Reaction - Following the announcement of V3.1, shares of Daily Interactive (300766.SZ), perceived as a DeepSeek concept stock, surged by 13.62% to close at 47.98 yuan per share [4][5]. Company Background - Daily Interactive, established in 2010, provides data intelligence products and services to commercial entities and government departments. There were rumors earlier this year that Daily Interactive held a 14.50% stake in DeepSeek's development team, but the company clarified that it does not own any shares in DeepSeek or its associated companies [6].

Group 1 - DeepSeek officially announced the release of version 3.1 on August 21, featuring significant upgrades including a hybrid reasoning architecture and improved response efficiency [1] - The new version utilizes UE8M0FP8Scale parameter precision and has made substantial adjustments to the tokenizer and chat template, showing clear differences from version 3 [1] - DeepSeek has adjusted its pricing for API interface calls, with input prices increasing to 0.5 yuan per million tokens for cache hits and 4 yuan for cache misses, while output prices rose to 12 yuan per million tokens [2] Group 2 - The foundational model of DeepSeek V3.1 underwent extensive retraining, adding a total of 840 billion tokens, and both the foundational and post-training models are available on Huggingface and Modao [4] - Following the announcement, shares of Daily Interactive (300766) surged, closing at 47.98 yuan per share with a daily increase of 13.62% [4] - Daily Interactive, established in 2010, provides data intelligence products and services, and there were rumors of its ownership stake in DeepSeek through its subsidiary, although the company later clarified it does not hold any equity in DeepSeek or its associated companies [7]

Group 1 - DeepSeek officially announced the release of version 3.1 on August 21, featuring significant upgrades including a hybrid reasoning architecture and improved response efficiency [1] - The new version utilizes UE8M0 FP8 Scale parameter precision, with major adjustments made to tokenizer 1 and chat template, marking a clear difference from DeepSeek version 3 [1] - Following the announcement, Daily Interactive (300766.SZ), considered a DeepSeek concept stock, surged by 13% during trading, closing at 47.98 CNY per share [1] Group 2 - Earlier this year, there were rumors that Daily Interactive held a 14.50% stake in the DeepSeek development team, Huansheng Quantitative, through its subsidiary Hangzhou Yingjing Technology Co., Ltd., which would allow it to participate in strategic decisions [2] - Daily Interactive clarified that it does not hold any equity in DeepSeek or Huansheng Technology, nor has it provided any corpus data to DeepSeek [2] - Daily Interactive is currently facing significant performance pressure, with a projected revenue of 470 million CNY for 2024, a year-on-year increase of 9.41%, but a net loss of 65 million CNY, a decline of 29.69% year-on-year [2]