llmsearchindex
LLMSearchIndex LLMSearchIndex is a library and search index that provides fully local, internet-scale web search for LLM RAG applications, trained on 203,169,792 web pages sourced from:
This index can be used as external context to significantly improve LLM responses without requiring external API calls at query time.
Demo
You can try out the search at: https://huggingface.co/spaces/zakerytclarke/llmsearchindex
Installation
pip install llmsearchindex
PyPI: https://pypi.org/project/llmsearchindex/
Model: https://huggingface.co/zakerytclarke/llmindex
Github: https://github.com/zakerytclarke/llmsearchindex
Example Usage:
from llmsearchindex import LLMIndex
# Initializes and downloads index
index = LLMIndex()
# Standard search (Fastest)
results = index.search("who invented sliced bread", top_k=5)
# High-precision search (Reranked)
results = index.search("who invented sliced bread", top_k=5, rerank=True)
for result in results:
print(result.get('url'))
print(result.get('text'))
print(result.get('snippet'))
print("==="*100)
System requirements
- ~6 GB RAM
- ~10 GB disk space
- CPU inference supported (GPU optional)
Architecture
flowchart LR
A[User Query] --> B(Embed Sentence Transformers all-MiniLM-L6)
B --> C(PCA: 384d → 64d)
C --> D(Binary Quantize)
X[HuggingFace FineWeb] --> G
Y[HuggingFace Wikipedia] --> G
D --> E{FAISS Index}
E --> G(Fetch Indexed Rows from HuggingFace Server)
G --> H{Rerank?}
H -->|Yes| I(Cosine Similarity)
H -->|No| J[Final Results]
I --> J
B -->I
Resources
Embeddings: https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2
FAISS Vector search: https://github.com/facebookresearch/faiss
Wikipedia: https://huggingface.co/datasets/wikimedia/wikipedia
FineWeb: https://huggingface.co/datasets/HuggingFaceFW/fineweb
License- MIT License
Metadata
Release files for llmsearchindex 1.2.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| llmsearchindex-1.2.0.tar.gz | 7.1 kB | Without provenance | Details for llmsearchindex-1.2.0.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| llmsearchindex-1.2.0-py3-none-any.whl | Python 3 | none | any | Details for llmsearchindex-1.2.0-py3-none-any.whl |
Total release size: 14.6 kB
Release files / llmsearchindex-1.2.0.tar.gz
| Download URL | llmsearchindex-1.2.0.tar.gz |
|---|---|
| Size | 7.1 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
d35987de7f74ebbdca6ee1f7dd48538168197511841e58a6bbc3b728a2f22a31
|
|
BLAKE2b-256 checksum How to use checksums |
6593142872cd315fee8300e229aa3b6cc0066b86ff0052bdae313a0236d30f25
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.12.3
|
Release files / llmsearchindex-1.2.0-py3-none-any.whl
| Download URL | llmsearchindex-1.2.0-py3-none-any.whl |
|---|---|
| Size | 7.5 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
d2fb4d4857e14cedd624867c16b8b69c14895bb4e000f75550fbe13026074e8f
|
|
BLAKE2b-256 checksum How to use checksums |
ae4e87bb3804787046a424e9a1ce695aa901e8bb9425877e4ea67900efcb4a6b
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.12.3
|
Security for llmsearchindex 1.2.0
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.