NVIDIA-NeMo / NVIDIA-NeMo/Guardrails
feature: Add non-LLM guardrails
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 7.2k
- Forks
- 842
- Avg merge
- 3d 1h
- Merged PRs (30d)
- 25
Description
Did you check the docs?
- I have read all the NeMo-Guardrails docs
Is your feature request related to a problem? Please describe.
I want to be able to use regex and XGB models for input and/or output rails.
While the pre-existing LLM based rails can perform the same types of input/output rails as regex and XGB models, there is a lot of overhead resources (eg. GPUs, memory) required. Regex and XGB based rails can provide a more lightweight solution, especially for simple content_safety and sensitive_data_detection tasks.
Describe the solution you'd like
I want to add regex and XGB based input/output rails.
Describe alternatives you've considered
There are currently no non-LLM based input/output rails.
Additional context
No response
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
No files, tests, or entry points are named in the issue. Start by reviewing the existing input/output rail documentation and implementations, then determine how regex and XGB models should be configured and invoked. Done means both model types support input and output rails for tasks such as content safety and sensitive data detection without requiring LLM resources.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- ai, security
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 35/100