注册了解独家消息、提示和版本发布

率先了解关于英特尔® 发行版 OpenVINO™ 工具套件的一切新内容。注册后，您可以获得抢先了解产品更新和发布信息、独家受邀参加网络研讨会和活动、培训和教程资源、竞赛公告以及其他突发新闻。

除非标为可选，否则所有字段均为必填。

公司电子邮件地址

请选择国家/地区

英特尔致力于为您提供优质、个性化的体验，您的数据帮助我们实现这一目标。

我同意英特尔按如下所述收集和使用我的个人信息。

我明确同意英特尔将我的个人数据传输到中国境外的英特尔美国总部，如《英特尔中国用户隐私声明补充》中所述。

提交此表单，即表示您确认自己已经年满 18 周岁，同意与英特尔分享您的个人信息以用于此业务需求。您还同意订阅通过电子邮件和电话及时获取英特尔® 技术和行业趋势的最新信息。您可以随时取消订阅。英特尔网站和通信内容遵循我们的隐私声明和使用条款。

OpenVINO™ 工具套件：一款可轻松实现“一次写入，处处部署”的开源 AI 工具套件。

OpenVINO 工具套件 2025.1 版通过新功能和性能增强，将您的 AI 部署提升到全新水平。该版本引入了对大语言模型 (LLM) 的持续改进、针对英特尔® 硬件的优化运行时，以及跨边缘、云和本地环境来高效部署 AI 的扩展功能。了解最新资讯，为您的 AI 项目开启全新可能性。

Easier Model Access and Conversion

Product	Details
New Model Support	New models supported: Phi-4-Mini, jina-clip-v1, and bce-embedding-base-v1.
OpenVINO™ Model Server Updates	Now supports vision language models (VLM), including Qwen2-VL, Phi-3.5-vision, and InternVL2.

GenAI and LLM Enhancements

Expanded model support and accelerated inference.

Feature	Details
CPU Plug-in Optimizations	Reduced the binary size through optimization of the CPU plug-in and removal of the General Matrix Multiplication (GEMM) kernel.
Kernel Optimization	Optimized new kernels for the GPU plug-in significantly boost the performance of long short-term memory (LSTM) models used in many applications, including speech recognition, language modeling, and time series forecasting.

More Portability and Performance

Develop once, deploy anywhere. OpenVINO toolkit enables developers to run AI at the edge, in the cloud, or locally.

Product	Details
Intel® Hardware Support	Support for the latest Intel® Core™ processors (Series 2), Intel® Core™ 3 processor N-series, and Intel® Processor N-series on Windows. Additional LLM performance optimizations on Intel® Core™ Ultra 200H series processors for improved second token latency on Windows and Linux. Enabled NPU acceleration for text generation in OpenVINO™ Runtime and OpenVINO model server to support the power-efficient deployment of VLM models on NPUs for AI PC use cases with low concurrency.
GenAI API Enhancements	OpenVINO™ GenAI now includes image-to-image and inpainting features for transformer-based pipelines, such as Flux.1 and Stable Diffusion 3 models, enhancing their ability to generate more realistic content. Preview: Token Eviction is implemented in OpenVINO GenAI to reduce the memory consumption of a key-value (KV) cache by eliminating unimportant tokens. This Token Eviction implementation is beneficial for tasks where a long sequence is generated, such as chatbots and code generation
Back End Integration	Preview: The new OpenVINO toolkit back end for ExecuTorch enables accelerated inference and improved performance on Intel hardware, including CPUs, GPUs, and NPUs.
Paged Attention and Continuous Batching Updates	Enhanced LLM performance and efficient resource use with the implementation of paged attention and continuous batching by default in the GPU plug-in.