# vLLM v0.4.0.post1 — Release & Upgrade Intelligence
- **Product:** vLLM (https://vllm.ai)
- **Version:** v0.4.0.post1
- **Released:** Apr 02, 2024
- **Impact Score:** 5/10
- **Breaking Changes:** Not verified
- **Canonical Analysis:** https://changelogs.us/vllm/updates/v0.4.0.post1/

## Executive Summary (TL;DR)
vLLM v0.4.0.post1. Source highlight: v0.4.0 lacks support for sm70/75 support. We did a hotfix for it.

## New Features
- [Misc] Add support for new autogptq checkpoint_format
- [Hardware][Intel] Add CPU inference backend

## Improvements
- [Kernel] Layernorm performance optimization

## Bug Fixes
- [Misc] Minor fixes in requirements.txt
- [Misc] Fix Benchmark TTFT Calculation for Chat Completions
- [HotFix] [CI/Build] Minor fix for CPU backend CI
- [Frontend][Bugfix] allow using the default middleware with a root path
- [Doc] Fix vLLMEngine Doc Page
- [CI/Build] fix TORCH_CUDA_ARCH_LIST in wheel build