New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

KV cache for faster generation #197

Closed

carmocca opened this issue Apr 27, 2023 · 0 comments · Fixed by #266

Labels

enhancement inference

Member

carmocca commented Apr 27, 2023

Reference: https://www.dipkumar.dev/becoming-the-unbeatable/posts/gpt-kvcache/
Whisper (MIT licensed) has a nice PyTorch-y implementation for this: https://github.com/openai/whisper/blob/c09a7ae299c4c34c5839a76380ae407e7d785914/whisper/model.py#L274-L305
nanoGPT doesn't implement this.

An argument against this is that it complicates the model code during training

Suggested by @Liyang90

carmocca added enhancement inference labels

carmocca mentioned this issue

KV cache for faster generation Lightning-AI/litgpt#21

Closed

gkroiz mentioned this issue

Adding cached KVs #266

Merged

lantiga closed this as completed in #266

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment