diff options
| author | Pierrick Hymbert <pierrick.hymbert@gmail.com> | 2024-03-01 12:39:06 +0100 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2024-03-01 13:39:06 +0200 |
| commit | 3ab8b3a92ede46df88bc5a2dfca3777de4a2b2b6 (patch) | |
| tree | b0232b770527ead35c5a59971a0802ded16e8c40 /llama.h | |
| parent | 9600d59e010c18f5872580a21734ea1bf1968d04 (diff) | |
llama : cleanup unused mmq flags (#5772)
* cleanup unused --no-mul-mat-q,-nommq, -mmq, --mul-mat-q, mul_mat_q
* remove: mul_mat_q in compare llama bench and usage
* update llama-bench
---------
Co-authored-by: slaren <slarengh@gmail.com>
Diffstat (limited to 'llama.h')
| -rw-r--r-- | llama.h | 1 |
1 files changed, 0 insertions, 1 deletions
@@ -255,7 +255,6 @@ extern "C" { enum ggml_type type_v; // data type for V cache // Keep the booleans together to avoid misalignment during copy-by-value. - bool mul_mat_q; // if true, use experimental mul_mat_q kernels (DEPRECATED - always true) bool logits_all; // the llama_eval() call computes all logits, not just the last one (DEPRECATED - set llama_batch.logits instead) bool embedding; // embedding mode only bool offload_kqv; // whether to offload the KQV ops (including the KV cache) to GPU |