From 4183a07ec4aefeba43260b5961d5409e7dc854fa Mon Sep 17 00:00:00 2001 From: Konrad Zawora Date: Tue, 8 Oct 2024 15:04:11 +0200 Subject: [PATCH] Update docs for 1.18.0 (#373) Updating docker links & version references --- Dockerfile.hpu | 2 +- README_GAUDI.md | 10 +++++----- docs/source/getting_started/gaudi-installation.rst | 8 ++++---- 3 files changed, 10 insertions(+), 10 deletions(-) diff --git a/Dockerfile.hpu b/Dockerfile.hpu index ab714cdac4670..50af19c5edb7b 100644 --- a/Dockerfile.hpu +++ b/Dockerfile.hpu @@ -1,4 +1,4 @@ -FROM vault.habana.ai/gaudi-docker/1.17.0/ubuntu22.04/habanalabs/pytorch-installer-2.3.1:latest +FROM vault.habana.ai/gaudi-docker/1.18.0/ubuntu22.04/habanalabs/pytorch-installer-2.4.1:latest COPY ./ /workspace/vllm diff --git a/README_GAUDI.md b/README_GAUDI.md index 04d620ff3bc60..773dac0d7982c 100644 --- a/README_GAUDI.md +++ b/README_GAUDI.md @@ -19,7 +19,7 @@ Requirements - OS: Ubuntu 22.04 LTS - Python: 3.10 - Intel Gaudi accelerator -- Intel Gaudi software version 1.17.0 +- Intel Gaudi software version 1.18.0 To verify that the Intel Gaudi software was correctly installed, run: @@ -45,8 +45,8 @@ for more details. Use the following commands to run a Docker image: ``` {.console} -$ docker pull vault.habana.ai/gaudi-docker/1.17.0/ubuntu22.04/habanalabs/pytorch-installer-2.3.1:latest -$ docker run -it --runtime=habana -e HABANA_VISIBLE_DEVICES=all -e OMPI_MCA_btl_vader_single_copy_mechanism=none --cap-add=sys_nice --net=host --ipc=host vault.habana.ai/gaudi-docker/1.17.0/ubuntu22.04/habanalabs/pytorch-installer-2.3.1:latest +$ docker pull vault.habana.ai/gaudi-docker/1.18.0/ubuntu22.04/habanalabs/pytorch-installer-2.4.1:latest +$ docker run -it --runtime=habana -e HABANA_VISIBLE_DEVICES=all -e OMPI_MCA_btl_vader_single_copy_mechanism=none --cap-add=sys_nice --net=host --ipc=host vault.habana.ai/gaudi-docker/1.18.0/ubuntu22.04/habanalabs/pytorch-installer-2.4.1:latest ``` Build and Install vLLM @@ -159,10 +159,10 @@ variable), and `--enforce-eager` flag. > [!WARNING] -> In 1.17.0, all modes utilizing `PT_HPU_LAZY_MODE=0` are highly +> In 1.18.0, all modes utilizing `PT_HPU_LAZY_MODE=0` are highly > experimental and should be only used for validating functional > correctness. Their performance will be improved in the next releases. -> For obtaining the best performance in 1.17.0, please use HPU Graphs, or +> For obtaining the best performance in 1.18.0, please use HPU Graphs, or > PyTorch lazy mode. Bucketing mechanism diff --git a/docs/source/getting_started/gaudi-installation.rst b/docs/source/getting_started/gaudi-installation.rst index 3a8b745a2a2f9..363fc921f944e 100644 --- a/docs/source/getting_started/gaudi-installation.rst +++ b/docs/source/getting_started/gaudi-installation.rst @@ -18,7 +18,7 @@ Requirements - OS: Ubuntu 22.04 LTS - Python: 3.10 - Intel Gaudi accelerator -- Intel Gaudi software version 1.17.0 +- Intel Gaudi software version 1.18.0 To verify that the Intel Gaudi software was correctly installed, run: @@ -45,8 +45,8 @@ Use the following commands to run a Docker image: .. code:: console - $ docker pull vault.habana.ai/gaudi-docker/1.17.0/ubuntu22.04/habanalabs/pytorch-installer-2.3.1:latest - $ docker run -it --runtime=habana -e HABANA_VISIBLE_DEVICES=all -e OMPI_MCA_btl_vader_single_copy_mechanism=none --cap-add=sys_nice --net=host --ipc=host vault.habana.ai/gaudi-docker/1.17.0/ubuntu22.04/habanalabs/pytorch-installer-2.3.1:latest + $ docker pull vault.habana.ai/gaudi-docker/1.18.0/ubuntu22.04/habanalabs/pytorch-installer-2.4.1:latest + $ docker run -it --runtime=habana -e HABANA_VISIBLE_DEVICES=all -e OMPI_MCA_btl_vader_single_copy_mechanism=none --cap-add=sys_nice --net=host --ipc=host vault.habana.ai/gaudi-docker/1.18.0/ubuntu22.04/habanalabs/pytorch-installer-2.4.1:latest Build and Install vLLM --------------------------- @@ -156,7 +156,7 @@ Currently in vLLM for HPU we support four execution modes, depending on selected - PyTorch lazy mode .. warning:: - In 1.17.0, all modes utilizing ``PT_HPU_LAZY_MODE=0`` are highly experimental and should be only used for validating functional correctness. Their performance will be improved in the next releases. For obtaining the best performance in 1.17.0, please use HPU Graphs, or PyTorch lazy mode. + In 1.18.0, all modes utilizing ``PT_HPU_LAZY_MODE=0`` are highly experimental and should be only used for validating functional correctness. Their performance will be improved in the next releases. For obtaining the best performance in 1.18.0, please use HPU Graphs, or PyTorch lazy mode. Bucketing mechanism