# vLLM v0.1.4 — Release & Upgrade Intelligence
- **Product:** vLLM (https://vllm.ai)
- **Version:** v0.1.4
- **Released:** Aug 25, 2023
- **Impact Score:** 7/10
- **Breaking Changes:** Source flagged
- **Canonical Analysis:** https://changelogs.us/vllm/updates/v0.1.4/

## Executive Summary (TL;DR)
vLLM v0.1.4. Source highlight: From now on, vLLM is published with pre-built CUDA binaries. Users don't have to compile the vLLM's CUDA kernels on their machine.

## Breaking Changes & Deprecations
- Fix for breaking changes in xformers 0.0.21 by @WoosukKwon in https://github.com/vllm-project/vllm/pull/834

## New Features
- add QWen-7b support
- add internlm model
- [Feature | CI] Added a github action to build wheels
- Add support for aquila
- Add compute capability 8.9 to default targets

## Bug Fixes
- Fix gibberish outputs of GPT-BigCode-based models
- [Fix] unwantted bias in InternLM Model
- Fix baichuan doc style
- Fix typo in tokenizer.py
- Fix typo in sampling_params.py
- Fix mqa is false case in gpt_bigcode

## Upgrade Command
```bash
python -m pip install 'vllm==0.1.4'
```