{"id":42480250,"url":"https://github.com/qdrant/qdrant-edge-demo","last_synced_at":"2026-01-28T10:39:28.328Z","repository":{"id":332312314,"uuid":"1121634851","full_name":"qdrant/qdrant-edge-demo","owner":"qdrant","description":"POC visual search with smart glasses and Qdrant Edge.","archived":false,"fork":false,"pushed_at":"2026-01-21T18:02:47.000Z","size":60606,"stargazers_count":36,"open_issues_count":1,"forks_count":2,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-01-22T05:36:17.446Z","etag":null,"topics":["ai-memory","ai-memory-system","edge","edge-ai","embedded-ai","vector-database","vector-search"],"latest_commit_sha":null,"homepage":"https://qdrant.tech/edge/ ","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/qdrant.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2025-12-23T09:54:13.000Z","updated_at":"2026-01-21T16:04:16.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/qdrant/qdrant-edge-demo","commit_stats":null,"previous_names":["qdrant/qdrant-edge-demo"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/qdrant/qdrant-edge-demo","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/qdrant%2Fqdrant-edge-demo","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/qdrant%2Fqdrant-edge-demo/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/qdrant%2Fqdrant-edge-demo/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/qdrant%2Fqdrant-edge-demo/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/qdrant","download_url":"https://codeload.github.com/qdrant/qdrant-edge-demo/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/qdrant%2Fqdrant-edge-demo/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":28844017,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-01-28T07:39:25.367Z","status":"ssl_error","status_checked_at":"2026-01-28T07:39:24.487Z","response_time":57,"last_error":"SSL_connect returned=1 errno=0 peeraddr=140.82.121.5:443 state=error: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["ai-memory","ai-memory-system","edge","edge-ai","embedded-ai","vector-database","vector-search"],"created_at":"2026-01-28T10:39:27.744Z","updated_at":"2026-01-28T10:39:28.323Z","avatar_url":"https://github.com/qdrant.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Smart Glasses x Qdrant Edge\n\nThis is a proof-of-concept for smart glasses that remember what they see and help you find your keys even without an internet connection.\n\nPowered by [Qdrant Edge](https://qdrant.tech/edge/).\n\n\u003cimg width=\"1272\" height=\"657\" alt=\"Qdrant Edge x Smart Glasses demo\" src=\"https://github.com/user-attachments/assets/4a46e4f1-12cd-41a4-99a7-9cf1ffa54a13\" /\u003e\n\n## How?\n\nThe system has two main parts: the glasses with Qdrant Edge and the Qdrant server.\n\n### On the device\n\nThe glasses use a [CLIP model](https://huggingface.co/Qdrant/clip-ViT-B-32-vision) to turn video frames into vectors. We compare frame similarity to skip redundant frames.\n\nVectors are stored locally in a mutable shard for immediate search. They're also pushed to a SQLite-backed queue that syncs to the server in the background. If the network drops, the queue holds onto data until connectivity returns.\n\n### Sync \u0026 Index\n\nBuilding the HNSW graph is CPU-heavy, so we offload it to the server. The glasses sync back the indexed data into an immutable shard.\n\nWe use this two-shard design to enable partial snapshots. Since the immutable shard is always an exact copy of the server, we can limit the snapshot download to only the new updates instead of the full database. Searches query both shards and merge results.\n\n\n## Try it out\n\nYou'll need [`uv`](https://docs.astral.sh/uv/getting-started/installation/) and [Docker](https://docs.docker.com/desktop/setup/install/mac-install/) installed.\n\n1. Start the server (handles the vector indexing):\n\n```bash\ndocker run -d -p 6333:6333 qdrant/qdrant\n```\n\n2. Install dependencies and download the CLIP models:\n\n```bash\nmake setup\n```\n\n3. Run the demo:\n\n```bash\nmake backend \u0026 make demo\n```\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fqdrant%2Fqdrant-edge-demo","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fqdrant%2Fqdrant-edge-demo","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fqdrant%2Fqdrant-edge-demo/lists"}