---
id: "lib-2026-10-02-inference-ml-llm-paas-vs"
type: "library"
title: "Зачем Авито нужна своя Inference-платформа для ML/LLM-сервисов: PaaS vs отдельное решение"
lang: "RU"
org: "Avito"
source_type: "Practitioner post"
date: "2026-10-02"
week: "2026-W40"
url: "https://nosov.ai/library/lib-2026-10-02-inference-ml-llm-paas-vs/"
original_url: "https://habr.com/ru/companies/avito/articles/1089128/"
tags: ["ai technology"]
---

# Зачем Авито нужна своя Inference-платформа для ML/LLM-сервисов: PaaS vs отдельное решение

## English summary

Why a web PaaS fails for LLM inference (GPU sharing, batching gives 5-10x utilisation difference); decision to build a separate inference platform on an open-source core.

Original: https://habr.com/ru/companies/avito/articles/1089128/