ver4a/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

kv-cache : add SWA support (#13194)

Browse source 
* kv-cache : prepare for SWA

ggml-ci

* kv-cache : initial iSWA implementation

ggml-ci

* kv-cache : rework error recovery logic

ggml-ci

* models : fix Phi-3 SWA parameters

ggml-ci

* model : adjust Granite to rope factor changes

ggml-ci

* server : check if context can do shifts

ggml-ci

* iswa : for now, always enable shifts (experiment)

ggml-ci

* kv-cache : simplify SWA logic

ggml-ci

* kv-cache : apply defrag when we fail to find slots for the batch

ggml-ci

* llama : update docs about llama_decode

ggml-ci

* kv-cache : update warning logs when no space for the batch is available

ggml-ci

* llama : add llama_kv_self_seq_pos_min()

* kv-cache : keep track of partial SWA computes and print warnings

* server : disallow use cases involving partial SWA context

ggml-ci

* llama : add param to control SWA cache size

ggml-ci

* minor : clean-up

ggml-ci
This commit is contained in:
Georgi Gerganov 2025-05-20 08:05:46 +03:00 • committed by GitHub
parent f0adb80bf7
commit e298d2fbd0
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
15 changed files with 1426 additions and 650 deletions
Download patch file Download diff file Expand all files Collapse all files

8
common/arg.cpp
View file

@ -1445,6 +1445,14 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
params.n_keep = value;
}
));
add_opt(common_arg(
{"--swa-full"},
string_format("use full-size SWA cache (default: %s)\n"
"[(more info)](https://github.com/ggml-org/llama.cpp/pull/13194#issuecomment-2868343055)", params.swa_full ? "true" : "false"),
[](common_params & params) {
params.swa_full = true;
}
));
add_opt(common_arg(
{"--no-context-shift"},
string_format("disables context shift on infinite text generation (default: %s)", params.ctx_shift ? "disabled" : "enabled"),