ggml-org/llama.cpp

server: avoid full prompt eval when 'prompt >= ctx'

クローズ

#6,855 opened on 2024/04/23

 (3 件のコメント) (0 件のリアクション) (0 人の担当者)C++ (21,764 件のフォーク)batch import
enhancementgood first issue

Repository metrics

Stars
 (124,134 個のスター)
PR merge metrics
 (平均マージ 6d 8h) (30d で 389 merged PRs)

説明

When using the server for multi-turn chat, soon or later the prompt is going to surpass the context size, the current approach truncate the prompt by half of the context size excluding n_keep:

https://github.com/ggerganov/llama.cpp/blob/192090bae47960f0d38d4967abe398a5d190057e/examples/server/server.cpp#L1969-L1983

By doing that, common_part is going to match only n_keep tokens (when cache_prompt: true):

https://github.com/ggerganov/llama.cpp/blob/192090bae47960f0d38d4967abe398a5d190057e/examples/server/server.cpp#L2011-L2016

Technically, this is not a full prompt eval, n_keep is not revaluated, but it would be better to avoid this if possible, specially because prompt eval is slow on CPU.

コントリビューターガイド