EngineX-Hygon/sglang
5
0
Fork 0
Code Issues Pull Requests Actions 7 Projects Releases Wiki Activity

Release 0.4.1.post3 - upload the config.json to PyPI (#2647)

Browse Source
This commit is contained in:
Lianmin Zheng
2024-12-29 14:25:53 -08:00
committed by GitHub
parent 1703d766d8
commit 03d5fbfd44
6 changed files with 17 additions and 16 deletions
Download Patch File Download Diff File Expand all files Collapse all files

11
benchmark/deepseek_v3/README.md
View File

@@ -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