# LocalAI v2.0.0 — Release & Upgrade Intelligence
- **Product:** LocalAI (https://localai.io)
- **Version:** v2.0.0
- **Released:** Dec 04, 2023
- **Impact Score:** 7/10
- **Breaking Changes:** Source flagged
- **Canonical Analysis:** https://changelogs.us/localai/updates/v2.0.0/

## Executive Summary (TL;DR)
LocalAI v2.0.0. Source highlight: :fire: add LLaVA support and GPT vision API, Multiple requests for llama.cpp, return JSON types

## Breaking Changes & Deprecations
- Breaking Changes 🛠 * :fire: add LLaVA support and GPT vision API, Multiple requests for llama.cpp, return JSON types by @mudler in https://github.com/mudler/LocalAI/pull/1254 * refactor: rename llama-stable to llama-ggml by @mudler in https://github.com/mudler/LocalAI/pull/1287

## New Features
- :fire: add LLaVA support and GPT vision API, Multiple requests for llama.cpp, return JSON types
- feat(certificates): add support for custom CA certificates
- feat(conda): conda environments
- refactor: move backends into the backends directory
- feat: allow to run parallel requests
- feat(transformers): add embeddings with Automodel

## Bug Fixes
- fix: respect OpenAI spec for response format
- fix: handle grpc and llama-cpp with REBUILD=true
- fix: propagate CMAKE_ARGS when building grpc
- fix(vall-e-x): correctly install reqs in environment
- Fixes the bug 1196
- fix(api/config): allow YAML config with .yml