# vLLM v0.1.3 — Release & Upgrade Intelligence
- **Product:** vLLM (https://vllm.ai)
- **Version:** v0.1.3
- **Released:** Aug 02, 2023
- **Impact Score:** 5/10
- **Breaking Changes:** Not verified
- **Canonical Analysis:** https://changelogs.us/vllm/updates/v0.1.3/

## Executive Summary (TL;DR)
vLLM v0.1.3. Source highlight: More model support: LLaMA 2, Falcon, GPT-J, Baichuan, etc.

## New Features
- Add trust-remote-code flag to handle remote tokenizers
- Add trust_remote_code arg to get_config
- [Model] Add support for GPT-J
- add vocab padding for LLama(Support WizardLM)
- [Doc] Add doc for running vLLM on the cloud
- Add support for baichuan

## Bug Fixes
- fix: only response [DONE] once when streaming response.
- [Fix] Change /generate response-type to json for non-streaming
- [Fix] Sort LLM outputs by request ID before return
- fix: freeze pydantic to v1
- Fix handling of special tokens in decoding.
- Fix the `KeyError` when loading bloom-based models

## Upgrade Command
```bash
python -m pip install 'vllm==0.1.3'
```