tqcq/llama.cpp
0
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-06-26 11:45:21 +00:00
Code Issues Packages Projects Releases Wiki Activity

llama : use n_swa + n_ubatch cells for SWA cache (#13833)

Browse Source 
* llama : use n_swa + n_ubatch cells for SWA cache

ggml-ci

* llama : add warning about multi-sqeuence SWA contexts
This commit is contained in:
Georgi Gerganov
2025-05-31 15:57:44 +03:00
committed by GitHub
parent c7e0a2054b
commit 3600cc2886
6 changed files with 24 additions and 11 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
include/llama.h
View File

@ -366,6 +366,8 @@ extern "C" {
bool no_perf; // measure performance timings
bool op_offload; // offload host tensor operations to device
bool swa_full; // use full-size SWA cache (https://github.com/ggml-org/llama.cpp/pull/13194#issuecomment-2868343055)
// NOTE: setting to false when n_seq_max > 1 can cause bad performance in some cases
// ref: https://github.com/ggml-org/llama.cpp/pull/13845#issuecomment-2924800573
};
// model quantization parameters
@ -502,6 +504,7 @@ extern "C" {
LLAMA_API int32_t llama_model_n_layer (const struct llama_model * model);
LLAMA_API int32_t llama_model_n_head (const struct llama_model * model);
LLAMA_API int32_t llama_model_n_head_kv (const struct llama_model * model);
LLAMA_API int32_t llama_model_n_swa (const struct llama_model * model);
// Get the model's RoPE frequency scaling factor
LLAMA_API float llama_model_rope_freq_scale_train(const struct llama_model * model);