@@ -12,13 +12,13 @@ To install the latest stable release:
1212
1313``` shell
1414# CPU-only version
15- $ conda install -c pytorch -c conda-forge faiss-cpu=1.14.1
15+ $ conda install -c pytorch -c conda-forge faiss-cpu=1.14.2
1616
1717# GPU(+CPU) version
18- $ conda install -c pytorch -c nvidia -c conda-forge faiss-gpu=1.14.1
18+ $ conda install -c pytorch -c nvidia -c conda-forge faiss-gpu=1.14.2
1919
2020# GPU(+CPU) version with NVIDIA cuVS
21- $ conda install -c pytorch -c nvidia -c rapidsai -c conda-forge libnvjitlink faiss-gpu-cuvs=1.14.1
21+ $ conda install -c pytorch -c nvidia -c rapidsai -c conda-forge libnvjitlink faiss-gpu-cuvs=1.14.2
2222
2323# GPU(+CPU) version using AMD ROCm not yet available
2424```
@@ -34,15 +34,15 @@ If you prefer [Pixi](https://pixi.sh/latest/), you can use the same channels and
3434``` shell
3535# CPU-only version
3636$ pixi init -c pytorch -c conda-forge
37- $ pixi add faiss-cpu=1.14.1
37+ $ pixi add faiss-cpu=1.14.2
3838
3939# GPU(+CPU) version
4040$ pixi init -c pytorch -c nvidia -c conda-forge
41- $ pixi add faiss-gpu=1.14.1
41+ $ pixi add faiss-gpu=1.14.2
4242
4343# GPU(+CPU) version with NVIDIA cuVS
4444$ pixi init -c pytorch -c nvidia -c rapidsai -c conda-forge
45- $ pixi add libnvjitlink faiss-gpu-cuvs=1.14.1
45+ $ pixi add libnvjitlink faiss-gpu-cuvs=1.14.2
4646```
4747
4848Pixi resolves packages from the listed Conda channels, so the same channel requirements described above still apply.
@@ -54,7 +54,7 @@ Nightly pre-release packages can be installed as follows:
5454$ conda install -c pytorch/label/nightly -c conda-forge faiss-cpu
5555
5656# GPU(+CPU) version
57- $ conda install -c pytorch/label/nightly -c nvidia -c conda-forge faiss-gpu=1.14.1
57+ $ conda install -c pytorch/label/nightly -c nvidia -c conda-forge faiss-gpu=1.14.2
5858
5959# GPU(+CPU) version with NVIDIA cuVS (package built with CUDA 13.2)
6060conda install -c pytorch -c rapidsai -c rapidsai-nightly -c conda-forge -c nvidia pytorch/label/nightly::faiss-gpu-cuvs ' cuda-version=13.2'
0 commit comments