# LocalAI v2.16.0 — Release & Upgrade Intelligence
- **Product:** LocalAI (https://localai.io)
- **Version:** v2.16.0
- **Released:** May 24, 2024
- **Impact Score:** 5/10
- **Breaking Changes:** Not verified
- **Canonical Analysis:** https://changelogs.us/localai/updates/v2.16.0/

## Executive Summary (TL;DR)
LocalAI v2.16.0. Source highlight: Welcome to LocalAI's latest update!

## New Features
- feat: migrate python backends from conda to uv
- feat: create bash library to handle install/run/test of python backends
- feat(grammar): support models with specific construct
- feat(ui): display number of available models for installation
- feat: auto select llama-cpp cpu variant
- feat(llama.cpp): add `flash_attention` and `no_kv_offloading`

## Improvements
- refactor: Minor improvements to BackendConfigLoader

## Bug Fixes
- build: do not specify a BUILD_ID by default
- fix: add missing openvino/optimum/etc libraries for Intel, fixes #2289
- add setuptools for openvino
- fix: add setuptools to all requirements-intel.txt files for python backends
- ci: correctly build p2p in GO_TAGS
- ci: generate specific image for intel builds