# Ollama releases version 0.40.17-rc0

> The open‑source AI runtime adds GPU memory mapping, concurrent generation, and a smoother ChatGPT CLI restart.

Oossa · 2026-11-10 · https://oossa.com/en/ollama-releases-version-0-40-17-rc0

Ollama announced version 0.40.17‑rc0. The update adds mmap support for Vulkan GPUs, letting the model use GPU memory more efficiently. It also enables concurrent text generation in the MLX backend and improves how the ChatGPT command‑line interface restores after a crash. The llama.cpp library was upgraded to commit b12257, and a new contributor made their first pull request.

## The facts

- Version 0.40.17‑rc0 released on 2026‑11‑10
- Adds Vulkan GPU mmap and concurrent MLX generation

## Why it matters

Developers can run larger models faster on compatible GPUs, reducing latency for local AI applications.

## Sources & references

1. [ollama/ollama v0.40.17](https://github.com/ollama/ollama/releases/tag/v0.40.17-rc0) – Ollama, 2026-11-10

Last updated: 2026-11-10