CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-10-27 08:21:30 +00:00
Code Issues Packages Projects Releases Wiki Activity

llama: use FA + max. GPU layers by default (#15434)

Browse Source 
* llama: use max. GPU layers by default, auto -fa

* ggml-backend: abort instead of segfault
This commit is contained in:
Johannes Gäßler
2025-08-30 16:32:10 +02:00
committed by GitHub
parent 38ad381f9f
commit e81b8e4b7f
19 changed files with 235 additions and 72 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
examples/diffusion/diffusion-cli.cpp
View File

@@ -564,7 +564,7 @@ int main(int argc, char ** argv) {
ctx_params.n_ctx = params.n_ctx;
ctx_params.n_batch = params.n_batch;
ctx_params.n_ubatch = params.n_ubatch;
ctx_params.flash_attn = params.flash_attn;
ctx_params.flash_attn_type = params.flash_attn_type;
ctx_params.no_perf = params.no_perf;
ctx_params.type_k = params.cache_type_k;
ctx_params.type_v = params.cache_type_v;