Mistral AI NewsJun 8, 2026, 9:02 AM

Heaps do lie: debugging a memory leak in vLLM

Original: Engineering Heaps do lie: debugging a memory leak in vLLM. January 21, 2026 By Mathis Felardos

Mistral AI explains how it traced a vLLM memory leak beyond the heap into UCX mmap hooks.

Mistral AI published an engineering deep dive on a memory leak found during vLLM disaggregated serving tests. The leak appeared only with a specific stack involving Mistral Medium 3.1, NIXL, UCX, graph compilation, and P/D disaggregation, with RSS growing steadily despite heap profilers looking normal. The team used pmap, BPFtrace, and targeted GDB automation to trace the issue to UCX mmap hooks and applied configuration fixes plus a vLLM patch.

想看英文原文 / 完整內容?

前往 Mistral AI News 原文 →

摘要由 AI 整理,以原文為準。