Mirrors/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2025-01-08 11:46:53 +01:00
Code Issues Packages Projects Releases Wiki Activity

llama : fix vocab_only logic when GPU is enabled

Browse Source
This commit is contained in:
Georgi Gerganov 2023-09-03 10:39:15 +03:00
parent 99161230c4
commit 323a9d3b8c
No known key found for this signature in database
GPG Key ID: 449E073F9DC10735
1 changed files with 24 additions and 24 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
llama.cpp
View File

@ -5547,7 +5547,6 @@ struct llama_context * llama_new_context_with_model(
} }
#endif #endif
} }
}
#ifdef GGML_USE_METAL #ifdef GGML_USE_METAL
if (params.n_gpu_layers > 0) { if (params.n_gpu_layers > 0) {
@ -5584,6 +5583,7 @@ struct llama_context * llama_new_context_with_model(
#undef LLAMA_METAL_CHECK_BUF #undef LLAMA_METAL_CHECK_BUF
} }
#endif #endif
}
#ifdef GGML_USE_MPI #ifdef GGML_USE_MPI
ctx->ctx_mpi = ggml_mpi_init(); ctx->ctx_mpi = ggml_mpi_init();