Skip to content

ROCm support of Ultra-Fast and Cheaper Long-Context LLM Inference

License

Notifications You must be signed in to change notification settings

EmbeddedLLM/LMCache

 
 
lmcache logo

💡 What is LMCache?

LMCache lets LLMs prefill each text only once. By storing the KV caches of all reusable texts, LMCache can reuse the KV caches of any reused text (not necessarily prefix) in any serving engine instance. It thus reduces prefill delay, i.e., time to first token (TTFT), as well as saves the precious GPU cycles.

By combining LMCache with vLLM, LMCaches achieves 3-10x delay savings and GPU cycle reduction in many LLM use cases, including multi-round QA and RAG.

Try LMCache with pre-built vllm docker images here.

🚀 Performance snapshot

image

💻 Quickstart

CUDA

LMCache provides the integration to the latest vLLM (0.6.2). To install LMCache, use the following command:

# requires python >= 3.10 and nvcc >= 12.1
pip install lmcache lmcache_vllm

LMCache has the same interface as vLLM (both online serving and offline inference). To use the online serving, you can start an OpenAI API-compatible vLLM server with LMCache via:

lmcache_vllm serve lmsys/longchat-7b-16k --gpu-memory-utilization 0.8

To use vLLM's offline inference with LMCache, just simply add lmcache_vllm before the import to vLLM components. For example

import lmcache_vllm.vllm as vllm
from lmcache_vllm.vllm import LLM 

More detailed documentation will be available soon.

ROCm

Prerequisite:

  • Installed vLLM 0.6.1.post2
  • Installed Pytorch 2.6.0+rocm6.2

To install LMCache, use the following command:

# requires python >= 3.10 and rocm >= 6.2
git clone https://github.com/EmbeddedLLM/torchac_rocm.git
cd torchac_rocm
# you might need to run this command twice to install if it fails the first time
HCC_AMDGPU_TARGET=gfx90a python3 setup.py develop
cd ..
git clone https://github.com/EmbeddedLLM/LMCache.git
python3 setup.py develop

- Sharing KV cache across multiple vLLM instances

LMCache supports sharing KV across different vLLM instances by the lmcache.server module. Here is a quick guide:

# Start lmcache server
lmcache_server localhost 65432

Then, start two vLLM instances with the LMCache config file

wget https://raw.githubusercontent.com/LMCache/LMCache/refs/heads/dev/examples/example.yaml

# start the first vLLM instance
LMCACHE_CONFIG_FILE=example.yaml CUDA_VISIBLE_DEVICES=0 lmcache_vllm serve lmsys/longchat-7b-16k --gpu-memory-utilization 0.8 --port 8000

# start the second vLLM instance
LMCACHE_CONFIG_FILE=example.yaml CUDA_VISIBLE_DEVICES=1 lmcache_vllm serve lmsys/longchat-7b-16k --gpu-memory-utilization 0.8 --port 8001

- What's next

We also provide multiple docker-based demos at 🔗LMCache-demos repo. The demos cover the following use cases:

  • Share KV caches across multiple serving engines (🔗link)
  • Loading non-prefix KV caches for RAG (🔗link)

Interested in Connecting?

Fill out the interest form and our team will reach out to you! https://forms.gle/mQfQDUXbKfp2St1z7

🛣️ Incoming Milestones

  • First release of LMCache
  • Support installation through pip install and integrate with latest vLLM
  • Stable support for non-prefix KV caches
  • User and developer documentation

📖 Blogs and documentations

Our blog posts and documentations are available online

Citation

If you use LMCache for your research, please cite our papers:

@inproceedings{liu2024cachegen,
  title={Cachegen: Kv cache compression and streaming for fast large language model serving},
  author={Liu, Yuhan and Li, Hanchen and Cheng, Yihua and Ray, Siddhant and Huang, Yuyang and Zhang, Qizheng and Du, Kuntai and Yao, Jiayi and Lu, Shan and Ananthanarayanan, Ganesh and others},
  booktitle={Proceedings of the ACM SIGCOMM 2024 Conference},
  pages={38--56},
  year={2024}
}

@article{cheng2024large,
  title={Do Large Language Models Need a Content Delivery Network?},
  author={Cheng, Yihua and Du, Kuntai and Yao, Jiayi and Jiang, Junchen},
  journal={arXiv preprint arXiv:2409.13761},
  year={2024}
}

@article{yao2024cacheblend,
  title={CacheBlend: Fast Large Language Model Serving with Cached Knowledge Fusion},
  author={Yao, Jiayi and Li, Hanchen and Liu, Yuhan and Ray, Siddhant and Cheng, Yihua and Zhang, Qizheng and Du, Kuntai and Lu, Shan and Jiang, Junchen},
  journal={arXiv preprint arXiv:2405.16444},
  year={2024}
}

About

ROCm support of Ultra-Fast and Cheaper Long-Context LLM Inference

Resources

License

Code of conduct

Security policy

Stars

Watchers

Forks

Packages

No packages published

Languages

  • Python 96.2%
  • Shell 3.8%