deepseekmirror/DeepSeek-V3
1
0
Fork 0
mirror of https://github.com/deepseek-ai/DeepSeek-V3.git synced 2025-05-21 17:56:45 -04:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity

State the support is in PyTorch backend.

Browse Source
This commit is contained in:
Bo Li 2025-05-21 11:07:08 +08:00 committed by GitHub
parent 643e176ed3
commit 91987eddb9
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
README.md
View File

@ -322,7 +322,7 @@ For comprehensive step-by-step instructions on running DeepSeek-V3 with LMDeploy
### 6.4 Inference with TRT-LLM (recommended)
[TensorRT-LLM](https://github.com/NVIDIA/TensorRT-LLM) now supports the DeepSeek-V3 model, offering features like FP8/[FP4](https://huggingface.co/nvidia/DeepSeek-R1-FP4) precision, Tensor Parallelism/Expert Parallelism for MoE/Data Parallelism for attention, Multi-Token Prediction etc. You can refer to this [README](https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/models/core/deepseek_v3/README.md) page for detailed instructions on usage, as well as this [blog](https://github.com/NVIDIA/TensorRT-LLM/blob/main/docs/source/blogs/Best_perf_practice_on_DeepSeek-R1_in_TensorRT-LLM.md) page for guidance on how to get best performance.
[TensorRT-LLM](https://github.com/NVIDIA/TensorRT-LLM) now supports the DeepSeek-V3 model in its re-architected [PyTorch backend](https://nvidia.github.io/TensorRT-LLM/torch.html), offering features like FP8/[FP4](https://huggingface.co/nvidia/DeepSeek-R1-FP4) precision, Tensor Parallelism/Expert Parallelism for MoE/Data Parallelism for attention, Multi-Token Prediction etc. You can refer to this [README](https://github.com/NVIDIA/TensorRT-LLM/blob/main/examples/models/core/deepseek_v3/README.md) page for detailed instructions on usage, as well as this [blog](https://github.com/NVIDIA/TensorRT-LLM/blob/main/docs/source/blogs/Best_perf_practice_on_DeepSeek-R1_in_TensorRT-LLM.md) page for guidance on how to get best performance.
### 6.5 Inference with vLLM (recommended)