From 03d5fbfd441ad1b03feb66a4d1cbb03088399eff Mon Sep 17 00:00:00 2001 From: Lianmin Zheng Date: Sun, 29 Dec 2024 14:25:53 -0800 Subject: [PATCH] Release 0.4.1.post3 - upload the config.json to PyPI (#2647) --- benchmark/deepseek_v3/README.md | 11 ++++++----- docker/Dockerfile.rocm | 2 +- docs/developer/setup_github_runner.md | 4 ++-- docs/start/install.md | 10 +++++----- python/pyproject.toml | 4 ++-- python/sglang/version.py | 2 +- 6 files changed, 17 insertions(+), 16 deletions(-) diff --git a/benchmark/deepseek_v3/README.md b/benchmark/deepseek_v3/README.md index 59ef8fb19b..b876ba1334 100644 --- a/benchmark/deepseek_v3/README.md +++ b/benchmark/deepseek_v3/README.md @@ -1,8 +1,6 @@ -# SGLang v0.4.1 - DeepSeek V3 Support +# DeepSeek V3 Support -We're excited to announce [SGLang v0.4.1](https://github.com/sgl-project/sglang/releases/tag/v0.4.1), which now supports [DeepSeek V3](https://huggingface.co/deepseek-ai/DeepSeek-V3-Base) - currently the strongest open-source LLM, even surpassing GPT-4o. - -The SGLang and DeepSeek teams worked together to get DeepSeek V3 FP8 running on NVIDIA and AMD GPU **from day one**. We've also supported MLA optimization and DP attention before, making SGLang one of the best open-source LLM engines for running DeepSeek models. +The SGLang and DeepSeek teams worked together to get DeepSeek V3 FP8 running on NVIDIA and AMD GPUs **from day one**. SGLang also has supported [MLA optimization](https://lmsys.org/blog/2024-09-04-sglang-v0-3/#deepseek-multi-head-latent-attention-mla-throughput-optimizations) and [DP attention](https://lmsys.org/blog/2024-12-04-sglang-v0-4/#data-parallelism-attention-for-deepseek-models), making SGLang one of the best open-source LLM engines for running DeepSeek models. Special thanks to Meituan's Search & Recommend Platform Team and Baseten's Model Performance Team for implementing the model, and DataCrunch for providing GPU resources. @@ -20,17 +18,20 @@ If you encounter errors when starting the server, ensure the weights have finish docker run --gpus all --shm-size 32g -p 30000:30000 -v ~/.cache/huggingface:/root/.cache/huggingface --ipc=host lmsysorg/sglang:latest \ python3 -m sglang.launch_server --model deepseek-ai/DeepSeek-V3 --tp 8 --trust-remote-code --port 30000 ``` + For high QPS scenarios, add the `--enable-dp-attention` argument to boost throughput. ### Using pip ```bash # Installation -pip install "sglang[all]==0.4.1.post2" --find-links https://flashinfer.ai/whl/cu124/torch2.4/flashinfer +pip install "sglang[all]>=0.4.1.post3" --find-links https://flashinfer.ai/whl/cu124/torch2.4/flashinfer # Launch python3 -m sglang.launch_server --model deepseek-ai/DeepSeek-V3 --tp 8 --trust-remote-code ``` +For high QPS scenarios, add the `--enable-dp-attention` argument to boost throughput. + ### Example with OpenAI API ```python3 diff --git a/docker/Dockerfile.rocm b/docker/Dockerfile.rocm index 8de4c40313..c416a4b1b2 100644 --- a/docker/Dockerfile.rocm +++ b/docker/Dockerfile.rocm @@ -1,5 +1,5 @@ # Usage (to build SGLang ROCm docker image): -# docker build --build-arg SGL_BRANCH=v0.4.1.post2 -t v0.4.1.post2-rocm620 -f Dockerfile.rocm . +# docker build --build-arg SGL_BRANCH=v0.4.1.post3 -t v0.4.1.post3-rocm620 -f Dockerfile.rocm . # default base image ARG BASE_IMAGE="rocm/vllm-dev:20241022" diff --git a/docs/developer/setup_github_runner.md b/docs/developer/setup_github_runner.md index 5f82f7153d..3c5589d0eb 100644 --- a/docs/developer/setup_github_runner.md +++ b/docs/developer/setup_github_runner.md @@ -11,9 +11,9 @@ docker pull nvidia/cuda:12.1.1-devel-ubuntu22.04 # Nvidia docker run --shm-size 128g -it -v /tmp/huggingface:/hf_home --gpus all nvidia/cuda:12.1.1-devel-ubuntu22.04 /bin/bash # AMD -docker run --rm --device=/dev/kfd --device=/dev/dri --group-add video --shm-size 128g -it -v /tmp/huggingface:/hf_home lmsysorg/sglang:v0.4.1.post2-rocm620 /bin/bash +docker run --rm --device=/dev/kfd --device=/dev/dri --group-add video --shm-size 128g -it -v /tmp/huggingface:/hf_home lmsysorg/sglang:v0.4.1.post3-rocm620 /bin/bash # AMD just the last 2 GPUs -docker run --rm --device=/dev/kfd --device=/dev/dri/renderD176 --device=/dev/dri/renderD184 --group-add video --shm-size 128g -it -v /tmp/huggingface:/hf_home lmsysorg/sglang:v0.4.1.post2-rocm620 /bin/bash +docker run --rm --device=/dev/kfd --device=/dev/dri/renderD176 --device=/dev/dri/renderD184 --group-add video --shm-size 128g -it -v /tmp/huggingface:/hf_home lmsysorg/sglang:v0.4.1.post3-rocm620 /bin/bash ``` ### Step 2: Configure the runner by `config.sh` diff --git a/docs/start/install.md b/docs/start/install.md index 9bd905e554..c297d10730 100644 --- a/docs/start/install.md +++ b/docs/start/install.md @@ -13,7 +13,7 @@ Note: Please check the [FlashInfer installation doc](https://docs.flashinfer.ai/ ## Method 2: From source ``` # Use the last release branch -git clone -b v0.4.1.post2 https://github.com/sgl-project/sglang.git +git clone -b v0.4.1.post3 https://github.com/sgl-project/sglang.git cd sglang pip install --upgrade pip @@ -26,7 +26,7 @@ Note: To AMD ROCm system with Instinct/MI GPUs, do following instead: ``` # Use the last release branch -git clone -b v0.4.1.post2 https://github.com/sgl-project/sglang.git +git clone -b v0.4.1.post3 https://github.com/sgl-project/sglang.git cd sglang pip install --upgrade pip @@ -51,7 +51,7 @@ docker run --gpus all \ Note: To AMD ROCm system with Instinct/MI GPUs, it is recommended to use `docker/Dockerfile.rocm` to build images, example and usage as below: ```bash -docker build --build-arg SGL_BRANCH=v0.4.1.post2 -t v0.4.1.post2-rocm620 -f Dockerfile.rocm . +docker build --build-arg SGL_BRANCH=v0.4.1.post3 -t v0.4.1.post3-rocm620 -f Dockerfile.rocm . alias drun='docker run -it --rm --network=host --device=/dev/kfd --device=/dev/dri --ipc=host \ --shm-size 16G --group-add video --cap-add=SYS_PTRACE --security-opt seccomp=unconfined \ @@ -60,11 +60,11 @@ alias drun='docker run -it --rm --network=host --device=/dev/kfd --device=/dev/d drun -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=" \ - v0.4.1.post2-rocm620 \ + v0.4.1.post3-rocm620 \ python3 -m sglang.launch_server --model-path meta-llama/Llama-3.1-8B-Instruct --host 0.0.0.0 --port 30000 # Till flashinfer backend available, --attention-backend triton --sampling-backend pytorch are set by default -drun v0.4.1.post2-rocm620 python3 -m sglang.bench_one_batch --batch-size 32 --input 1024 --output 128 --model amd/Meta-Llama-3.1-8B-Instruct-FP8-KV --tp 8 --quantization fp8 +drun v0.4.1.post3-rocm620 python3 -m sglang.bench_one_batch --batch-size 32 --input 1024 --output 128 --model amd/Meta-Llama-3.1-8B-Instruct-FP8-KV --tp 8 --quantization fp8 ``` ## Method 4: Using docker compose diff --git a/python/pyproject.toml b/python/pyproject.toml index 2fe240e284..c51e21f50e 100644 --- a/python/pyproject.toml +++ b/python/pyproject.toml @@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta" [project] name = "sglang" -version = "0.4.1.post2" +version = "0.4.1.post3" description = "SGLang is yet another fast serving framework for large language models and vision language models." readme = "README.md" requires-python = ">=3.8" @@ -61,7 +61,7 @@ dev_hpu = ["sglang[all_hpu]", "sglang[test]"] "Bug Tracker" = "https://github.com/sgl-project/sglang/issues" [tool.setuptools.package-data] -"sglang" = ["srt/layers/fused_moe_triton/configs/*.json"] +"sglang" = ["srt/layers/moe/fused_moe_triton/configs/*.json", "srt/layers/quantization/configs/*.json"] [tool.setuptools.packages.find] exclude = [ diff --git a/python/sglang/version.py b/python/sglang/version.py index d05c9f06ef..5ab5956b57 100644 --- a/python/sglang/version.py +++ b/python/sglang/version.py @@ -1 +1 @@ -__version__ = "0.4.1.post2" +__version__ = "0.4.1.post3"