mirror of
https://github.com/ggml-org/llama.cpp.git
synced 2026-11-11 14:22:09 +01:00
ggml-cpu: FA split across kv for faster TG (#19209)
* ggml-cpu: split across kv for faster TG * simplify sinks application * add ref impl
This commit is contained in:
@@ -24,6 +24,9 @@ struct ggml_compute_params {
|
||||
void * wdata;
|
||||
|
||||
struct ggml_threadpool * threadpool;
|
||||
|
||||
// use reference implementation
|
||||
bool use_ref;
|
||||
};
|
||||
|
||||
|
||||
|
||||
Reference in New Issue
Block a user