ggml-org/llama.cpp
server: avoid full prompt eval when 'prompt >= ctx'
Fermée
#6 855 ouverte le 23 avr. 2024
enhancementgood first issue
Métriques du dépôt
- Stars
- (124 134 étoiles)
- Métriques de merge PR
- (Merge moyen 6j 8h) (389 PRs mergées en 30 j)
Description
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:
By doing that, common_part is going to match only n_keep tokens (when cache_prompt: true):
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.