LazyHuman10 commited on
Commit
688d628
·
1 Parent(s): e8d932a

Fix HF README metadata

Browse files
Files changed (1) hide show
  1. README.md +19 -40
README.md CHANGED
@@ -1,40 +1,19 @@
1
- # Plexi API
2
-
3
- FastAPI RAG backend deployed on HuggingFace Spaces.
4
-
5
- ## What It Does
6
-
7
- - Loads the pre-built LlamaIndex vector store from the `plexi-materials` GitHub repo at startup
8
- - Exposes three endpoints consumed by the Cloudflare Worker middleman
9
-
10
- ## Endpoints
11
-
12
- | Method | Path | Purpose |
13
- |---|---|---|
14
- | `GET` | `/health` | Liveness probe — used by keep-alive GitHub Actions |
15
- | `GET` | `/manifest` | Proxies + caches `manifest.json` from the materials repo |
16
- | `POST` | `/retrieve` | Embeds query, searches index, returns scoped top-k chunks |
17
-
18
- ## Local Development
19
-
20
- ```bash
21
- pip install -r requirements.txt
22
- uvicorn main:app --reload --port 7860
23
- ```
24
-
25
- Visit `http://localhost:7860/docs` for the interactive API docs.
26
-
27
- ## Environment Variables
28
-
29
- | Variable | Default | Purpose |
30
- |---|---|---|
31
- | `MATERIALS_REPO` | `KunalGupta25/plexi-materials` | GitHub repo with study materials |
32
- | `MANIFEST_BRANCH` | `main` | Branch that holds `manifest.json` and `index/` |
33
- | `ALLOWED_ORIGINS` | (Cloudflare Pages URL) | CORS allowed origins |
34
-
35
- ## Deploy to HuggingFace Spaces
36
-
37
- 1. Create a new Space with **Docker** SDK
38
- 2. Push this folder as the Space repo
39
- 3. Set environment variables in the Space settings
40
- 4. HF will build and run the Dockerfile automatically
 
1
+ title: Plexi API
2
+ emoji: 🧠
3
+ colorFrom: blue
4
+ colorTo: cyan
5
+ sdk: docker
6
+ pinned: false
7
+ license: mit
8
+ ---
9
+
10
+ # Plexi API (The Brain)
11
+
12
+ This is the backend API for Plexi, serving as "The Brain" for the application.
13
+ It is built with FastAPI and runs on Hugging Face Spaces using a custom Docker container.
14
+
15
+ ## Features
16
+ - Document parsing and chunking
17
+ - High-performance vector embeddings (Sentence Transformers)
18
+ - Context retrieval for RAG (Retrieval-Augmented Generation) pipelines
19
+ - Manifest generation for available study materials