Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
An NVIDIA AI Workbench example project for Retrieval Augmented Generation (RAG)
| Date | Stars |
|---|---|
| 2026-07-31 | 369 |
| 2026-08-06 | 370 |
Today
+1 stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
# An Easy Button for Retrieval Augmented Generation (RAG) [](https://ngc.nvidia.com/open-ai-workbench/aHR0cHM6Ly9naXRodWIuY29tL05WSURJQS93b3JrYmVuY2gtZXhhbXBsZS1oeWJyaWQtcmFn)
This chat application works interchangeably with build.nvidia.com API endpoints, downloadable NVIDIA NIM containers, and Hugging Face TGI on NVIDIA GPUs. It makes testing RAG with NVIDIA endpoints, self-hosted NIMs, and third party models simple.
<!-- Banner Image -->
<img src="https://developer-blogs.nvidia.com/wp-content/uploads/2024/07/rag-representation.jpg" width="100%">
> **Note**
> This app runs in [NVIDIA AI Workbench](https://docs.nvidia.com/ai-workbench/user-guide/latest/overview/introduction.html). It's a free, lightweight developer platform that you can run on your own systems to get up and running with complex AI applications and workloads in a short amount of time.
> You may want to [**fork**](https://docs.github.com/en/pull-requests/collaborating-with-pull-requests/working-with-forks/fork-a-repo#forking-a-repository) this repository into your own account before proceeding. Otherwise you won't be able to fully push any changes you make because this NVIDIA-owned repository is **read-only**.
<br>
*Navigating the README*: [Project Overview](#project-overview) | [Quickstart](#quickstart) | [Troubleshooting](#troubleshooting) | [Advanced Tutorials](#advanced-tutorials) | [License](#license)
*Other Resources*: [:arrow_down: Download AI Workbench](https://www.nvidia.com/en-us/deep-learning-ai/solutions/data-science/workbench/) | [:book: User Guide](https://docs.nvidia.com/ai-workbench/) |[:open_file_folder: Other Projects](https://docs.nvidia.com/ai-workbench/user-guide/latest/quickstart/example-projects.html) | [:rotating_light: Support Thread](https://forums.developer.nvidia.com/t/support-workbench-example-project-hybrid-rag/288565)
## Project Overview
This is an [NVIDIA AI Workbench](https://www.nvidia.com/en-us/deep-learning-ai/solutions/data-science/workbench/) project for developing a [Retrieval Augmented Generation](https://blogs.nvidia.com/blog/what-is-retrieval-augmented-generation/) application with a customizable Gradio Chat app. It lets you:
* Embed your documents into a locally running vector database.
* Run inference **locally** on a Hugging Face TGI server, in the **cloud** using NVIDIA inference endpoints, or using **microservices** via [NIMs](https://www.nvidia.com/en-us/ai/):
* 4-bit, 8-bit, and no quantization options are supported for locally running models served by TGI.
* Other models may be specified to run locally using their Hugging Face tag.
* Any downloadable NIM model can be used as a microservice, runnable via docker/podman compose in AI Workbench
### Table 1 Default Supported Models by Inference Mode
<details>
<summary>
<b>Expand this section for a full table on all supported models by inference mode.</b>
</summary>
| Model | Local ([TGI](https://huggingface.co/docs/text-generation-inference/en/index)) | Cloud ([NVIDIA API Catalog](https://build.nvidia.com/explore/discover)) | Microservices ([NVIDIA NIMs](https://www.nvidia.com/en-us/ai/)) |
| -------- | ----------------------------- | --------------- | ------------------------------ |
| Llama3-ChatQA-1.5-8B | Y | Y | * |
| Llama3-ChatQA-1.5-70B | | Y | * |
| Nemotron-Mini-4B | | Y | * |
| Nemotron-4-340B-Instruct | | Y | * |
| Mistral-NeMo 12B Instruct | | Y | * |
| Mistral-7B-Instruct-v0.1 | Y (gated) | | * |
| Mistral-7B-Instruct-v0.2 | Y (gated) | Y | * |
Excerpt of 24,377 characters
Read on GitHubWould you bet a product on this? Bounded 0–100 and slow moving.
matched fp:6402e6f3b67b3588, desc:retrieval-augmented generation, desc:retrieval augmented