lmcache--lmcache/examples/kv_cache_reuse

Examples of KV cache reusing and sharing with vLLM + LMCache

LMCache should be able to reduce the generation time of the second and following calls.

We have examples for the following types of KV cache sharing and reusing:

• KV cache reusing with local storage backends: local_backends/
• KV cache reusing with remote storage backends: remote_backends/
• KV cache sharing across different vLLM instances: share_across_instances/