# vLLM v0.3.3 — Release & Upgrade Intelligence
- **Product:** vLLM (https://vllm.ai)
- **Version:** v0.3.3
- **Released:** Mar 01, 2024
- **Impact Score:** 5/10
- **Breaking Changes:** Not verified
- **Canonical Analysis:** https://changelogs.us/vllm/updates/v0.3.3/

## Executive Summary (TL;DR)
vLLM v0.3.3. Source highlight: Performance optimization and LoRA support for Gemma

## New Features
- Added early stopping to completion APIs
- Add LogProbs for Chat Completions in OpenAI
- Add LoRA support for Gemma
- Add Support for 2/3/8-bit GPTQ Quantization Models
- add cache_config's info to prometheus metrics.
- Add guided decoding for OpenAI API server

## Improvements
- Performance optimization and LoRA support for Gemma
- Performance optimization for MoE kernel

## Bug Fixes
- [FIX] Fix issue #2904
- Fix nvcc not found in vllm-openai image
- [Fix] Fix assertion on Mistral YaRN model len
- Fix using CuPy for eager mode
- Fix stablelm
- fix `get_ip` error in pure ipv6 environment

## Upgrade Command
```bash
python -m pip install 'vllm==0.3.3'
```