vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

llama : fix session saving/loading

Browse source
This commit is contained in:
Georgi Gerganov 2023-09-29 15:47:21 +03:00
parent c71bf2c45c
commit b0670db34f
No known key found for this signature in database
GPG key ID: 449E073F9DC10735
3 changed files with 65 additions and 38 deletions
Download patch file Download diff file Expand all files Collapse all files

2
llama.h
View file

@ -42,7 +42,7 @@
#define LLAMA_FILE_MAGIC_GGSN 0x6767736eu // 'ggsn'
#define LLAMA_SESSION_MAGIC LLAMA_FILE_MAGIC_GGSN
#define LLAMA_SESSION_VERSION 1
#define LLAMA_SESSION_VERSION 2
#if defined(GGML_USE_CUBLAS) || defined(GGML_USE_CLBLAST) || defined(GGML_USE_METAL)
// Defined when llama.cpp is compiled with support for offloading model layers to GPU.