SilverTorch: 索引即模型 — 推荐系统的一种新型检索范式

We’re introducing SilverTorch, a reimagining of recommendation systems that unifies all retrieval components for user generated content under a unified architecture.
我们推出了SilverTorch，这是对推荐系统的重新构想，它将用户生成内容的所有检索组件统一在一个统一的架构下。
SilverTorch shows up to 23.7x higher throughput compared to the state-of-the-art approaches. It’s also showing 20.9x more compute cost efficiency compared to a CPU-based solution while also improving accuracy.
与最先进的方法相比，SilverTorch的吞吐量提高了高达23.7倍。与基于CPU的解决方案相比，它的计算成本效率也提高了20.9倍，同时还提高了准确性。
Our research paper, “SilverTorch: A Unified Model-based System to Democratize Large-Scale Recommendation on GPUs,” accepted to the full paper track at SIGIR 2026, contains full technical details.
我们的研究论文“SilverTorch: A Unified Model-based System to Democratize Large-Scale Recommendation on GPUs”已被 SIGIR 2026 全文轨道录用，其中包含了完整的技术细节。

The retrieval system within industry recommendation systems have consisted of microservices stitched together, with neural networks inconsistently integrated. Our recommendation can scale to serve people across multiple platforms. Retrieval is responsible for narrowing from millions of pieces of content (e.g., reels and photos) down to thousands before passing them to ranking systems, all in less than 100 milliseconds.

行业推荐系统中的检索系统一直由拼接在一起的微服务组成，神经网络的集成并不一致。我们的推荐系统可以扩展以服务多个平台上的用户。检索负责在将数百万条内容（例如短视频和照片）传递给排序系统之前，将其缩小到数千条，所有这一切都在不到 100 毫秒的时间内完成。

However, the microservice based design had hard constraints on model complexity and the number of candidates evaluated, ultimately creating a ceiling on the quality of recommendations that people on our platforms see.

然而，基于微服务的设计对模型复杂度和评估的候选数量有严格的限制，最终为我们平台上用户看到的推荐质量设定了上限。

To break through this ceiling, we’ve fully reimagined our retrieval ecosystem into a unified model-based system – SilverTorch.

为了突破这一上限，我们将检索生态系统全面重构为一个统一的基于模型的系统——SilverTorch。

SilverTorch operates under a new paradigm we call Index as Model. We’ve built our retrieval system as a single neural network and now express different microservices as model modules within this integrat...