# Building Pinterest's VLM Serving Stack on NVIDIA Dynamo

DevFeed: [Building Pinterest's VLM Serving Stack on NVIDIA Dynamo](<https://devfeed.tech/articles/building-pinterest-s-vlm-serving-stack-on-nvidia-dynamo-1229.md>)

Original publisher: [Read original article](<https://medium.com/pinterest-engineering/building-pinterests-vlm-serving-stack-on-nvidia-dynamo-0dce6e93d0f3?source=rss----4c5a5f6279b6---4>)

Author: Pinterest Engineering

Published: 2026-09-10T23:08:16Z

Content type: article

Language: en

Sources: [Pinterest Engineering Blog - Medium](<https://devfeed.tech/sources/pinterest-engineering-blog-medium.md>)

Topics: [vlm](<https://devfeed.tech/topics/vlm.md>), [Inference](<https://devfeed.tech/topics/inference.md>), [model-deployment](<https://devfeed.tech/topics/model-deployment.md>), [Orchestration](<https://devfeed.tech/topics/orchestration.md>), [Latency](<https://devfeed.tech/topics/latency.md>), [Caching](<https://devfeed.tech/topics/caching.md>), [Open Source Models & Datasets](<https://devfeed.tech/topics/open-source-models-datasets.md>)

Tags: [blackwell](<https://devfeed.tech/tags/blackwell.md>), [cache](<https://devfeed.tech/tags/cache.md>), [dynamo](<https://devfeed.tech/tags/dynamo.md>), [embeddings](<https://devfeed.tech/tags/embeddings.md>), [engineering](<https://devfeed.tech/tags/engineering.md>), [inference](<https://devfeed.tech/tags/inference.md>), [latency](<https://devfeed.tech/tags/latency.md>), [llm](<https://devfeed.tech/tags/llm.md>), [multimodal](<https://devfeed.tech/tags/multimodal.md>), [multimodal-ai](<https://devfeed.tech/tags/multimodal-ai.md>), [nvidia](<https://devfeed.tech/tags/nvidia.md>), [nvidia-blackwell](<https://devfeed.tech/tags/nvidia-blackwell.md>), [open-source](<https://devfeed.tech/tags/open-source.md>), [orchestration](<https://devfeed.tech/tags/orchestration.md>), [performance](<https://devfeed.tech/tags/performance.md>), [pinterest](<https://devfeed.tech/tags/pinterest.md>), [vllm](<https://devfeed.tech/tags/vllm.md>), [vlm](<https://devfeed.tech/tags/vlm.md>), [vlm-serving](<https://devfeed.tech/tags/vlm-serving.md>)

## AI overview

Pinterest describes its VLM serving stack built on NVIDIA Blackwell GPUs and NVIDIA Dynamo. The stack addresses multimodal inference demands such as image processing, variable prefill costs, KV-cache pressure, routing, and cache offloading.

## Source excerpt

Lei Pan | Senior Software Engineer; Salina Wu | Senior Software Engineer; Cristian Lopez | Software Engineer I; Guangtong Bai | Staff Software Engineer; Soam Acharya | Principal Engineer; Saurabh Vishwas Joshi | Principal Engineer; Chia-Wei Chen | Staff Software Engineer; Ambud Sharma | Principal Engineer Why VLM Serving Matters at Pinterest Pinterest is a visual search and discovery platform, so its AI systems must reason over both language and visual content. Vision-language models (VLMs), which can interpret images, compare visual candidates, and respond naturally to user intent, are becoming the foundation for the next generation of Pinterest experiences: Pinterest Assistant, hybrid search, multimodal reranking, content understanding, signal generation, content safety, and more. This direction also reflects Pinterest's broader strategy to customize open-source models to meet its product & scale needs. Pinterest Assistant is a standout example. This multi-turn conversational experience covers both user language and visual content. Serving it requires low-latency VLM inference over rich multimodal context as well as reworking Qwen3-VL with proprietary multimodal embeddings to cut runtime cost while improving performance. Serving VLMs, however, introduces more challenges compared to text-only LLM workloads. Requests may carry multiple images, require extra vision encoder computation, incur larger and more variable prefill cost, and create higher KV cache pressure. To support this new class of models & product experiences, we built Pinterest's VLM serving stack on top of NVIDIA Blackwell GPUs and NVIDIA Dynamo. Blackwell GPUs incorporate many architectural innovations that are uniquely positioned for today's most demanding AI workloads -- including higher BF16/FP8 compute throughput, increased memory bandwidth, and larger HBM memory capacity -- that enable dramatically higher performance for inference. Dynamo provides a distributed inference orchestration layer that g