Spaces:
Running on Zero
Running on Zero
|
Download README.md from akshit-Gupta/Lexi-memory-engine: direct link, hf CLI and curl.
- Browser
- Download file 894 Bytes
-
https://huggingface.co/spaces/akshit-Gupta/Lexi-memory-engine/resolve/main/README.md
- Command line
-
hf download hf://spaces/akshit-Gupta/Lexi-memory-engine/README.md
-
curl -L -o README.md https://huggingface.co/spaces/akshit-Gupta/Lexi-memory-engine/resolve/main/README.md
894 Bytes
| title: Lexi | |
| emoji: "🧠" | |
| colorFrom: green | |
| colorTo: blue | |
| sdk: gradio | |
| sdk_version: "6.5.1" | |
| python_version: "3.12.12" | |
| app_file: app.py | |
| pinned: false | |
| # Lexi | |
| Context-aware personal phonetic memory demonstration. | |
| This Space hosts the live Lexi application and the Evaluation Lab. The Evaluation Lab exposes only **Auto**. The uploaded CSV describes the user's journey and expected behavior; Lexi decides whether to resolve a case locally, with semantic ML retrieval, or with a rare Gemini ambiguity check. | |
| ## Secrets | |
| Set `GEMINI_API_KEY` in the Space Secrets to enable the optional Gemini ambiguity resolver. The app remains usable without the key. | |
| ## ZeroGPU | |
| The Space is designed for ZeroGPU. MiniLM semantic retrieval is invoked only when the local contextual path is insufficient. A lightweight local fallback keeps the app usable if semantic model initialization is unavailable. | |