Release llamafile v0.8.5

Mozilla-Ocho · May 25, 2024 · b79ecf4 · b79ecf4
1 parent 07e87bf
commit b79ecf4
Show file tree

Hide file tree

Showing 3 changed files with 11 additions and 3 deletions.
diff --git a/llamafile/cuda.bat b/llamafile/cuda.bat
@@ -5,7 +5,11 @@
 :: driver on a Windows system that has a CUDA-capable GPU installed.
 
 nvcc --shared ^
-     -arch=all-major ^
+     --use_fast_math ^
+     -gencode arch=compute_60,code=sm_60 ^
+     -gencode arch=compute_70,code=sm_70 ^
+     -gencode arch=compute_80,code=sm_80 ^
+     -gencode arch=compute_90,code=sm_90 ^
      --forward-unknown-to-host-compiler ^
      -Xcompiler="/nologo /EHsc /O2 /GR /MT" ^
      -DNDEBUG ^

diff --git a/llamafile/cuda.sh b/llamafile/cuda.sh
@@ -23,7 +23,11 @@ cd "$TMP"
 
 /usr/local/cuda/bin/nvcc \
   --shared \
-  -arch=all-major \
+  --use_fast_math \
+  -gencode arch=compute_60,code=sm_60 \
+  -gencode arch=compute_70,code=sm_70 \
+  -gencode arch=compute_80,code=sm_80 \
+  -gencode arch=compute_90,code=sm_90 \
   --forward-unknown-to-host-compiler \
   --compiler-options "-fPIC -O2" \
   -DNDEBUG \

diff --git a/llamafile/version.h b/llamafile/version.h
@@ -2,7 +2,7 @@
 
 #define LLAMAFILE_MAJOR 0
 #define LLAMAFILE_MINOR 8
-#define LLAMAFILE_PATCH 4
+#define LLAMAFILE_PATCH 5
 #define LLAMAFILE_VERSION \
     (100000000 * LLAMAFILE_MAJOR + 1000000 * LLAMAFILE_MINOR + LLAMAFILE_PATCH)