starred/ollama
1
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-04-23 17:29:54 +02:00
Code Issues Packages Projects Releases Wiki Activity

runner: Remove CGO engines, use llama-server exclusively for GGML models

Browse Source 
Remove the vendored GGML and llama.cpp backend, CGO runner, Go model
implementations, and sample.  llama-server (built from upstream llama.cpp via
FetchContent) is now the sole inference engine for GGUF-based models.
(Safetensor based models continue to run on the new MLX engine.)  This allows
us to more rapidly pick up new capabilities and fixes from llama.cpp as they
come out.

On windows this now requires recent AMD driver versions to support ROCm v7 as
llama.cpp currently does not support building against v6.
This commit is contained in:
Daniel Hiltgen
2026-03-25 16:59:18 -07:00
parent ff23dd343f
commit 56c735d871
902 changed files with 5888 additions and 423367 deletions
Download Patch File Download Diff File Expand all files Collapse all files

20
scripts/build_linux.sh
View File

@@ -59,16 +59,16 @@ fi
# buildx behavior changes for single vs. multiplatform
echo "Compressing linux tar bundles..."
if echo $PLATFORM | grep "," > /dev/null ; then
tar c -C ./dist/linux_arm64 --exclude cuda_jetpack5 --exclude cuda_jetpack6 . | zstd --ultra -22 -T0 >./dist/ollama-linux-arm64.tar.zst
tar c -C ./dist/linux_arm64 ./lib/ollama/cuda_jetpack5 | zstd --ultra -22 -T0 >./dist/ollama-linux-arm64-jetpack5.tar.zst
tar c -C ./dist/linux_arm64 ./lib/ollama/cuda_jetpack6 | zstd --ultra -22 -T0 >./dist/ollama-linux-arm64-jetpack6.tar.zst
tar c -C ./dist/linux_amd64 --exclude rocm . | zstd --ultra -22 -T0 >./dist/ollama-linux-amd64.tar.zst
tar c -C ./dist/linux_amd64 ./lib/ollama/rocm | zstd --ultra -22 -T0 >./dist/ollama-linux-amd64-rocm.tar.zst
tar c -C ./dist/linux_arm64 --exclude cuda_jetpack5 --exclude cuda_jetpack6 . | zstd -9 -T0 >./dist/ollama-linux-arm64.tar.zst
tar c -C ./dist/linux_arm64 ./lib/ollama/cuda_jetpack5 | zstd -9 -T0 >./dist/ollama-linux-arm64-jetpack5.tar.zst
tar c -C ./dist/linux_arm64 ./lib/ollama/cuda_jetpack6 | zstd -9 -T0 >./dist/ollama-linux-arm64-jetpack6.tar.zst
tar c -C ./dist/linux_amd64 --exclude rocm . | zstd -9 -T0 >./dist/ollama-linux-amd64.tar.zst
tar c -C ./dist/linux_amd64 ./lib/ollama/rocm | zstd -9 -T0 >./dist/ollama-linux-amd64-rocm.tar.zst
elif echo $PLATFORM | grep "arm64" > /dev/null ; then
tar c -C ./dist/ --exclude cuda_jetpack5 --exclude cuda_jetpack6 bin lib | zstd --ultra -22 -T0 >./dist/ollama-linux-arm64.tar.zst
tar c -C ./dist/ ./lib/ollama/cuda_jetpack5 | zstd --ultra -22 -T0 >./dist/ollama-linux-arm64-jetpack5.tar.zst
tar c -C ./dist/ ./lib/ollama/cuda_jetpack6 | zstd --ultra -22 -T0 >./dist/ollama-linux-arm64-jetpack6.tar.zst
tar c -C ./dist/ --exclude cuda_jetpack5 --exclude cuda_jetpack6 bin lib | zstd -9 -T0 >./dist/ollama-linux-arm64.tar.zst
tar c -C ./dist/ ./lib/ollama/cuda_jetpack5 | zstd -9 -T0 >./dist/ollama-linux-arm64-jetpack5.tar.zst
tar c -C ./dist/ ./lib/ollama/cuda_jetpack6 | zstd -9 -T0 >./dist/ollama-linux-arm64-jetpack6.tar.zst
elif echo $PLATFORM | grep "amd64" > /dev/null ; then
tar c -C ./dist/ --exclude rocm bin lib | zstd --ultra -22 -T0 >./dist/ollama-linux-amd64.tar.zst
tar c -C ./dist/ ./lib/ollama/rocm | zstd --ultra -22 -T0 >./dist/ollama-linux-amd64-rocm.tar.zst
tar c -C ./dist/ --exclude rocm bin lib | zstd -9 -T0 >./dist/ollama-linux-amd64.tar.zst
tar c -C ./dist/ ./lib/ollama/rocm | zstd -9 -T0 >./dist/ollama-linux-amd64-rocm.tar.zst
fi