root/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2024-12-27 03:44:35 +00:00
Code Issues Actions 4 Packages Projects Releases Wiki Activity

server : enable cache_prompt by default (#10501)

Browse Source 
ggml-ci
This commit is contained in:
Georgi Gerganov 2024-11-25 21:50:07 +02:00 committed by GitHub
parent 106964e3d2
commit 47f931c8f9
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
2 changed files with 3 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
examples/server/README.md
View File

@ -412,7 +412,7 @@ node index.js
`id_slot`: Assign the completion task to an specific slot. If is -1 the task will be assigned to a Idle slot. Default: `-1` `id_slot`: Assign the completion task to an specific slot. If is -1 the task will be assigned to a Idle slot. Default: `-1`
`cache_prompt`: Re-use KV cache from a previous request if possible. This way the common prefix does not have to be re-processed, only the suffix that differs between the requests. Because (depending on the backend) the logits are **not** guaranteed to be bit-for-bit identical for different batch sizes (prompt processing vs. token generation) enabling this option can cause nondeterministic results. Default: `false` `cache_prompt`: Re-use KV cache from a previous request if possible. This way the common prefix does not have to be re-processed, only the suffix that differs between the requests. Because (depending on the backend) the logits are **not** guaranteed to be bit-for-bit identical for different batch sizes (prompt processing vs. token generation) enabling this option can cause nondeterministic results. Default: `true`
`samplers`: The order the samplers should be applied in. An array of strings representing sampler type names. If a sampler is not set, it will not be used. If a sampler is specified more than once, it will be applied multiple times. Default: `["dry", "top_k", "typ_p", "top_p", "min_p", "xtc", "temperature"]` - these are all the available values. `samplers`: The order the samplers should be applied in. An array of strings representing sampler type names. If a sampler is not set, it will not be used. If a sampler is specified more than once, it will be applied multiple times. Default: `["dry", "top_k", "typ_p", "top_p", "min_p", "xtc", "temperature"]` - these are all the available values.

4
examples/server/server.cpp
View File

@ -111,7 +111,7 @@ struct server_static_file {
struct slot_params { struct slot_params {
bool stream = true; bool stream = true;
bool cache_prompt = false; // remember the prompt to avoid reprocessing all prompt bool cache_prompt = true; // remember the prompt to avoid reprocessing all prompt
int32_t n_keep = 0; // number of tokens to keep from initial prompt int32_t n_keep = 0; // number of tokens to keep from initial prompt
int32_t n_discard = 0; // number of tokens after n_keep that may be discarded when shifting context, 0 defaults to half int32_t n_discard = 0; // number of tokens after n_keep that may be discarded when shifting context, 0 defaults to half
@ -883,7 +883,7 @@ struct server_context {
} }
slot.params.stream = json_value(data, "stream", false); slot.params.stream = json_value(data, "stream", false);
slot.params.cache_prompt = json_value(data, "cache_prompt", false); slot.params.cache_prompt = json_value(data, "cache_prompt", true);
slot.params.n_predict = json_value(data, "n_predict", json_value(data, "max_tokens", defaults.n_predict)); slot.params.n_predict = json_value(data, "n_predict", json_value(data, "max_tokens", defaults.n_predict));
slot.params.n_indent = json_value(data, "n_indent", defaults.n_indent); slot.params.n_indent = json_value(data, "n_indent", defaults.n_indent);
slot.params.n_keep = json_value(data, "n_keep", defaults.n_keep); slot.params.n_keep = json_value(data, "n_keep", defaults.n_keep);