kanshan/TensorRT-LLMs
1
0
Fork 0
mirror of https://github.com/NVIDIA/TensorRT-LLM.git synced 2026-01-30 07:33:48 +08:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity
cbca6505ff
TensorRT-LLMs/tensorrt_llm/scaffolding/contrib
History
WeiHaocheng 8a994d879f
feat: fix erros on scaffolding README (#3899) 
Signed-off-by: fredw (generated by with_the_same_user script) <20514172+WeiHaocheng@users.noreply.github.com>
2025-04-29 10:15:06 +08:00
..
AsyncGeneration feat(part 2): Enhance the integrated robustness of scaffolding with __init__.py #3305 (#3731) 2025-04-24 18:47:03 +08:00
Dynasor feat(part 2): Enhance the integrated robustness of scaffolding with __init__.py #3305 (#3731) 2025-04-24 18:47:03 +08:00
__init__.py feat(part 2): Enhance the integrated robustness of scaffolding with __init__.py #3305 (#3731) 2025-04-24 18:47:03 +08:00
README.md feat: fix erros on scaffolding README (#3899) 2025-04-29 10:15:06 +08:00

README.md

Contrib Examples

We create this directory to store the community contributed projects.

Contributors can develop inference time compute methods with various Controller/Task/Worker.

We will continue to move some generic works on this directory back to the main code.

How to create a new project?

Just create a new directory in tensorrt_llm/scaffolding/contrib/ and add your code there.

How to make your code include Controller/Task/Worker can be reused by other projects?

Just add your Controller/Task/Worker to the __init__.py file of tensorrt_llm/scaffolding/contrib/.

How to show examples of your project?

Just add your example to the examples/scaffolding/contrib/ directory.

In summary, the part of the code you want to be imported by other users or projects should be put on tensorrt_llm/scaffolding/contrib/ directory and added to the __init__.py file. The code to run the project and show the results should be put on examples/scaffolding/contrib/ directory.