# vLLM v0.2.4 — Release & Upgrade Intelligence
- **Product:** vLLM (https://vllm.ai)
- **Version:** v0.2.4
- **Released:** Dec 11, 2023
- **Impact Score:** 5/10
- **Breaking Changes:** Not verified
- **Canonical Analysis:** https://changelogs.us/vllm/updates/v0.2.4/

## Executive Summary (TL;DR)
vLLM v0.2.4. Source highlight: Mixtral model support (officially from @mistralai)

## New Features
- add custom server params
- [Docker] Add cuda arch list as build option
- [Minor] Add comment on skipping rope caches
- Update README.md to add megablocks requirement for mixtral

## Bug Fixes
- Fix typo in adding_model.rst
- Fix quickstart.rst example
- fix typo in getenv call
- Fix Baichuan2-7B-Chat
- Fix for KeyError on Loading LLaMA
- [Minor] Fix code style for baichuan

## Upgrade Command
```bash
python -m pip install 'vllm==0.2.4'
```