Sift Video
Sift Video is a free, open-source multimodal search engine for finding moments in video through natural-language queries.
What Sift Video is
Sift Video is a free, open-source multimodal search engine for finding moments in video through natural-language queries. It searches spoken audio and visual content, then returns matching timestamps. Videos are processed offline: audio and frames become semantic embeddings indexed in Qdrant. You can download yt-dlp-compatible URLs while server-sent events report ingestion progress, or upload MP4, WebM, MKV, MOV, AVI, and OGV files up to 1 GB through the web interface. Results include video names, timestamps, match types, transcript excerpts, and relevance scores, with a link to open each video at the matching moment. Filters limit searches to audio, visual, or all matches, and searches can target one selected video. A REST API covers search, downloads, uploads, video listing and streaming, deletion, and health checks. Docker Compose instructions cover CPU deployment, with a separate NVIDIA GPU setup. The documented stack includes Whisper, MobileCLIP2, FFmpeg, and Qdrant. It is licensed under GNU GPLv3.
Good
- Searches both audio and visual content
- Natural-language queries return timestamps
- Supports API access and self-hosting
- Uploads common video formats up to 1 GB
- Offers CPU and NVIDIA GPU deployment
Watch out
- No hosted platform is listed
- Uploads are limited to 1 GB
- Requires Docker and Git setup
- NVIDIA mode requires host tooling
Who it suits
Sift Video suits people who need local, natural-language searching across video speech and imagery. Its Docker, Git, API, and optional NVIDIA GPU setup also fit technical teams building or hosting their own search service.
Verdict
Sift Video combines audio and visual timestamp search with a documented API and self-hosted deployment path. It is most suitable when local processing and technical control matter.
The facts
Each one read from Sift Video's own pages, with the date it was checked.
What it does
- API
- The REST API includes search, URL download progress, file upload, video listing and streaming, deletion, and a health endpoint.checked 8 Oct 2026
- Audience
- The README provides Docker, Git, and optional NVIDIA GPU setup instructions, and invites users to contribute issues or pull requests.checked 7 Oct 2026
- Audio and visual search
- The search interface supports all-match, audio-only, and visual-only filtering.checked 8 Oct 2026
- Deployment
- The project runs through Docker Compose and serves its web interface at localhost:8080.checked 8 Oct 2026
- File uploads
- The web UI accepts MP4, WebM, MKV, MOV, AVI, and OGV files up to 1 GB.checked 7 Oct 2026
- GPU support
- Separate NVIDIA GPU builds can use CUDA for Whisper transcription and ONNX Runtime inference, with NVIDIA Container Toolkit required on the host.checked 8 Oct 2026
- License
- The repository states that the project is licensed under GNU GPLv3.checked 8 Oct 2026
- Live ingestion
- The system accepts yt-dlp-compatible URLs and streams ingestion progress while downloading, extracting frames, transcribing with Whisper, and embedding content.checked 8 Oct 2026
- Local data handling
- The README describes a local Docker deployment with video and output directories and a Qdrant service; it does not state a security certification or compliance status.checked 7 Oct 2026
- Offline indexing
- Videos are processed offline, with audio and frames converted into semantic embeddings and indexed in Qdrant.checked 8 Oct 2026
- Processing
- Videos are processed offline, with audio and frames converted into semantic embeddings and indexed in Qdrant.checked 7 Oct 2026
- Purpose
- SIFT-Video searches video audio and visual content with natural-language queries and returns relevant timestamps.checked 7 Oct 2026
- Search filters
- Users can filter searches to audio, visual, or all matches and scope searches to one selected video.checked 7 Oct 2026
- Search results
- Results show video names, timestamps, match types, transcript excerpts, and relevance scores, and can open the built-in player at the matching moment.checked 8 Oct 2026
- Support and contributions
- The README welcomes suggestions, fixes, and improvements through issues or pull requests.checked 8 Oct 2026
- Technology
- The documented stack includes Whisper small.en for timestamped speech transcription and MobileCLIP2-S3-OpenCLIP for text and image embeddings.checked 8 Oct 2026
- Upload limits
- Local uploads support mp4, webm, mkv, mov, avi, and ogv files up to 1 GB.checked 8 Oct 2026
- URL ingestion
- The project can download yt-dlp-compatible URLs and stream ingestion progress through server-sent events.checked 7 Oct 2026
- What it does
- SIFT-Video is a multimodal semantic video search engine that searches video audio and visual content with natural-language queries and returns relevant timestamps.checked 8 Oct 2026
The company
- Maker
- The GitHub profile identifies the maker as Sourav Kumar.checked 8 Oct 2026
Alternatives to Sift Video
- 1
Deepgrip
Free planWebAPILinuxmacOS7.7From $89 a month - 2
FrameQuery
Free planWebAPImacOSWindows7.7From $19 a month - 3
VGG Image Annotator
Free planOpen sourceWebLinuxmacOSSelf-hosted7.7Free plan - 4
DISKO
Free planWebmacOS7.4From $12 a month - 5
Twelve Labs
Free planWebAPI7.3From $29 a month - 6
ViSearch
Free planWebAPIExtensionSelf-hosted7.3$0.09 one-time