# vLLM v0.5.1 — Release & Upgrade Intelligence
- **Product:** vLLM (https://vllm.ai)
- **Version:** v0.5.1
- **Released:** Jul 05, 2024
- **Impact Score:** 7/10
- **Breaking Changes:** Source flagged
- **Canonical Analysis:** https://changelogs.us/vllm/updates/v0.5.1/

## Executive Summary (TL;DR)
vLLM v0.5.1. Source highlight: vLLM now has pipeline parallelism! (#4412, #5408, #6115, #6120). You can now run the API server with `--pipeline-parallel-size`. This feature is in early stage, please let us know your feedback.

## Breaking Changes & Deprecations
- Notably, it has a **breaking change** that all VLM specific arguments are now removed from engine APIs so you no longer need to set it globally via CLI. However, you now only need to pass in `<image>` into the prompt instead of complicated prompt formatting. See more [here](https://docs.vllm.ai/en/latest/models/vlm.html#offline-batched-inference)
- [Doc] Add notice about breaking changes to VLMs by @DarkLight1337 in https://github.com/vllm-project/vllm/pull/5818

## New Features
- vLLM now has pipeline parallelism! (#4412, #5408, #6115, #6120). You can now run the API server with `--pipeline-parallel-size`. This feature is in early stage, please let us know your feedback.
- [CI/Build] Add `is_quant_method_supported` to control quantization test configurations
- Revert "[CI/Build] Add `is_quant_method_supported` to control quantization test configurations"
- [ci] Add AMD, Neuron, Intel tests for AWS CI and turn off default soft fail for GPU tests
- [misc] add hint for AttributeError
- [Frontend] Add "input speed" to tqdm postfix alongside output speed

## Improvements
- Enhancement to TPU support (#5292, #5878, #5850, #5831, #5855)
- Enhancement in distributed communication via shared memory (#5399)
- Latency enhancement in block manager (#5584)
- Enhancements to `compressed-tensors` supporting Marlin, W4A16 (#5435, #5385)
- Faster FP8 quantize kernel (#5396), FP8 on Ampere (#5975)
- Option to use FlashInfer for prefill, decode, and CUDA Graph for decode (#4628)

## Bug Fixes
- [Bugfix] Add device assertion to TorchSDPA
- [Bugfix] TYPE_CHECKING for MultiModalData
- [Doc] Update debug docs
- [Bugfix] Fix typo in scheduler.py (requeset -> request)
- [Bugfix] Fix wrong multi_modal_input format for CPU runner
- [Bugfix]if the content is started with ":"(response of ping), client should i…

## Upgrade Command
```bash
python -m pip install 'vllm==0.5.1'
```