diff options
| author | stduhpf <stephduh@live.fr> | 2024-01-16 12:04:32 +0100 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2024-01-16 13:04:32 +0200 |
| commit | e0324285a569d0583cf2f4a07a2402221ee25f58 (patch) | |
| tree | 8372c01c23d94190463f02c5e643142995aa7f84 /common/common.h | |
| parent | 3e5ca7931c68152e4ec18d126e9c832dd84914c8 (diff) | |
speculative : threading options (#4959)
* speculative: expose draft threading
* fix usage format
* accept -td and -tbd args
* speculative: revert default behavior when -td is unspecified
* fix trailing whitespace
Diffstat (limited to 'common/common.h')
| -rw-r--r-- | common/common.h | 2 |
1 files changed, 2 insertions, 0 deletions
diff --git a/common/common.h b/common/common.h index 09646824..1f43e628 100644 --- a/common/common.h +++ b/common/common.h @@ -46,7 +46,9 @@ struct gpt_params { uint32_t seed = -1; // RNG seed int32_t n_threads = get_num_physical_cores(); + int32_t n_threads_draft = -1; int32_t n_threads_batch = -1; // number of threads to use for batch processing (-1 = use n_threads) + int32_t n_threads_batch_draft = -1; int32_t n_predict = -1; // new tokens to predict int32_t n_ctx = 512; // context size int32_t n_batch = 512; // batch size for prompt processing (must be >=32 to use BLAS) |