Skip to content

Conversation

pathorn
Copy link

@pathorn pathorn commented Dec 7, 2023

Tested on mistralai/Mistral-7B-Instruct-v0.1

Also tested with llama and it still seems to work.

batch.prefill_next_token_indices = None
batch.max_seqlen = batch.max_seqlen + 1

batch.kv_cache_usage = CACHE_MANAGER.usage()

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I'm a bit worried that this call might cause perf drop. Can you try with and without it.

@pathorn pathorn changed the base branch from main to kv-cache December 8, 2023 01:47
@NikolaBorisov NikolaBorisov changed the title Add support for MIstral Add support for Mistral Dec 15, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants