Merged
Conversation
cristibleotiu
approved these changes
Apr 8, 2026
What changed: - Added extensions/utils/faiss_vectordb.py — FAISS adapter with GPU auto-detection - Rewired base_doc_emb_serving.py to use FaissVectorDB instead of HNSWVectorDB/docarray - Replaced vectordb with faiss-cpu>=1.7.4 in requirements files - Removed HNSWLIB_NO_NATIVE env from Dockerfiles - Cleaned up stale vectordb cache patterns from .gitignore Why: - jina-ai/vectordb is archived, unmaintained since 2023, and broken on Python 3.12+ - FAISS scored 4.70/5.00 in comparison benchmark — perfect recall, fastest search, minimal deps Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
What changed: - Removed faiss-cpu>=1.7.4 from requirements.txt and .devcontainer/requirements.txt Why: - FAISS is already provided by the base Docker images (faiss-cpu in CPU, faiss-gpu in GPU) - No need to duplicate the dependency in edge_node
What changed: - Added FastAPI test plugin for verifying FaissVectorDB adapter - Endpoints: status, add_docs, search, list_contexts, reset_context - Uses hash-based deterministic embeddings for testing Why: - Validates the FAISS replacement end-to-end on a live edge node - All 9 tests passed against dvi-1 via ngrok HTTP tunnel
What changed:
- All Dockerfiles now use ratio1/base_edge_node_amd64_cpu:latest (Python 3.13, PyTorch 2.9.1)
- Base image is parameterized via BASE_IMAGE build arg for CPU/GPU flexibility
- Cleaned up IPFS install (parameterized Kubo version, single layer, cleanup)
- Removed old apt workarounds (base image already provides wget, tar, ca-certificates)
- Added .github/workflows/build_gpu.yml for GPU image builds
Why:
- Old base image (Python 3.10, PyTorch 2.3.1) is outdated
- GPU builds reuse same Dockerfiles with --build-arg BASE_IMAGE=ratio1/base_edge_node_amd64_gpu:latest
- GPU images pushed to ratio1/edge_node_gpu:{devnet,testnet,mainnet}
toderian
force-pushed
the
vectordb-replacement
branch
from
3093753 to
f2b15a0
Compare
cristibleotiu
approved these changes
Apr 8, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
No description provided.