登录新浪财经APP 搜索【信披】查看更多考评等级 出品：新浪财经创投Plus 编辑整理：shu 基于公开数据不完全统计，2025年7月新增登记的私募股权、创业投资基金管理人共16家，较6月激增77.8%，达到2024年7月的四倍；新增备案私募股权投 资基金、创业投资基金分别为130只、245只，合计同比增长7.1%，环比下降3.4%。 国内一级股权投资市场共发生552起融资事件，同比、环比分别增长5.1%、11.7%；披露总融资额约717.56亿元，较2024年7月提高142.0%，与2025年6月相比 涨幅超100%；平均单笔融资额接近1.3亿元，创下近7个月内最高点。 新浪财经创投Plus本月聚焦市场活跃度较高的3家机构，围绕其投资节奏、阶段、行业偏好以及被投项目等方面展开解读。 7月投资事件节奏强势反弹 跟投MOSS大模型创始人公司 从投资阶段来看，奇绩创坛作为孵化器重点关注早期阶段投资，其参投的天使轮项目约占比55.6%、种子轮占比16.7%、A轮占比22.2%。从所关注的行业赛 道来看，奇绩创坛本月的投资一半投向人工智能行业，其中智能机器人、AI通用应用获投公司合计约占三分之二。由国内首个对话式大型 ...

Core Insights - Beijing Qingcheng Jizhi Technology Co., Ltd. has completed over 100 million yuan in financing, just six months after its previous round, led by notable industry players and supported by several strategic investors [1][8] - The company, established in December 2023, focuses on developing intelligent computing system software, acting as a crucial link between hardware computing power and AI model training and application needs [3] Financing and Investment - The latest financing round attracted significant investment from Lenovo Capital, a state-owned venture capital platform in Shanghai, and other strategic investors [1] - Qingcheng Jizhi previously completed an angel round in March 2024 and a second round in January 2025, with participation from various investment institutions [7] Product Development and Technology - The company has developed the "Bagua Furnace" high-performance model training system in collaboration with Tsinghua University, achieving significant acceleration in training tasks for dense models and mixed expert models [5] - The "Chitu" inference engine, optimized for domestic computing power, offers low latency, high throughput, and low memory usage, reducing the computational threshold for running the full version of DeepSeek by 75% [7] Strategic Goals - The CEO of Qingcheng Jizhi stated that the recent financing will enhance the company's capabilities in product development, ecosystem construction, and market expansion [8] - The company aims to contribute to the development of the domestic computing industry and the widespread application of AI technology [8]

Core Insights - Qingcheng Jizhi, a leading player in the AI infrastructure sector, has successfully completed a new financing round exceeding 100 million yuan, marking its third round of funding within a year [1][3] - The investment was led by a prominent industry player, with participation from various influential capital sources in the computing power industry, indicating strong market interest in AI infrastructure [1][4] - The CEO of Qingcheng Jizhi highlighted that the funding will enhance the company's capabilities in product development, ecosystem building, and market expansion [2][4] Company Overview - Qingcheng Jizhi was established in December 2023 and focuses on developing intelligent computing system software, acting as a crucial link between intelligent computing and applications [3][6] - The company is led by a team from Tsinghua University, with its CEO being a PhD graduate from the same institution [3][6] Financing Strategy - The company has adopted a "small steps, quick runs" financing strategy due to the rapid changes in the AI infrastructure industry and the high demand for cash flow to support product development [4][6] - This strategy has attracted multiple investors who are optimistic about the company's technological capabilities [4][5] Technological Advancements - Qingcheng Jizhi's software solutions provide a full-stack approach to optimize underlying computing power, enhancing model training and inference efficiency while reducing costs for AI application development [6][7] - The company has developed the "Bagualu" high-performance model training system, which has shown significant acceleration in training tasks on large-scale domestic computing clusters [6][7] - The "Chitu" inference engine has been optimized for domestic computing power, achieving low latency and high throughput, thus supporting diverse application scenarios [6][7] Market Demand - The demand for computing power in the AI market has shifted from primarily training large models to a significant increase in inference power requirements [7][8] - Qingcheng Jizhi is addressing the diverse needs of clients, from small enterprises seeking cost-effective solutions to large organizations requiring comprehensive computing power solutions [8]

Core Insights - Tsinghua-affiliated AI company Qingcheng Jizhi has recently completed over 100 million yuan in financing, less than six months after its previous round [1] - The latest funding round was led by a well-known industry player, with participation from various notable investment institutions [1] Company Overview - Qingcheng Jizhi focuses on developing intelligent computing system software, acting as a crucial bridge between intelligent computing and AI applications [1] - The company's software efficiently links underlying hardware computing power with upper-layer AI model training, inference, and application needs, facilitating seamless collaboration among different hardware devices [1] Technological Advancements - The company has achieved significant improvements in training efficiency for domestic chips, which will enhance the utilization and performance of domestic computing resources while reducing costs for enterprises [2] - Qingcheng Jizhi's "Bagualu" high-performance large model training system has been validated on multiple large-scale domestic computing clusters, showing notable acceleration in training tasks for dense models and mixed expert models [2] - The "Chitu" inference engine, developed by Qingcheng Jizhi, is optimized for domestic computing, offering low latency, high throughput, and low memory usage, thus meeting diverse intelligent computing needs [3] Strategic Collaborations - The company collaborates with Tsinghua University to optimize key aspects of model algorithms and system design, enhancing the overall efficiency of large model training [2] - The open-source Chitu inference engine project aims to accelerate the establishment of a complete ecosystem comprising domestic intelligent computing chips, system software, and large models [3]