ggml-org/llama.cpp
server: avoid full prompt eval when 'prompt >= ctx'
Chiusa
#6855 aperta il 23 apr 2024
enhancementgood first issue
Metriche repository
- Star
- (124.043 stelle)
- Metriche merge PR
- (Merge medio 6g 8h) (389 PR mergiate in 30 g)
Descrizione
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.