Gemini can now natively embed video, so I built sub-second video search
Gemini Embedding 2 can project raw video directly into a 768-dimensional vector space alongside text. No transcription, no frame captioning, no intermediate text. A query like "green car cutting me off" is directly comparable to a 30-second video clip at the vector level. I used this to build a CLI that indexes hours of footage into ChromaDB, then searches it with natural language and auto-trims the matching clip. Demo video on the GitHub README. Indexing costs ~$2.50/hr of footage. Still-frame detection skips idle chunks, so security camera / sentry mode footage is much cheaper.
In plain words
A command-line tool that indexes video footage by converting it directly into a searchable vector space using Gemini's native video embedding. Users can search hours of video with natural language queries like "green car cutting me off" and automatically retrieve matching clips. The tool stores indexed footage in ChromaDB and skips idle frames to reduce costs, making it practical for security camera and dashcam footage. Indexing costs approximately $2.50 per hour of video.
written from the facts on this page · September 2026
Does the same job
all alternatives →
- NWNeedle: We Distilled Gemini Tool Calling into a 26M ModelMay 2026 · github.com · ▲776
Hey HN, Henry here from Cactus. We open-sourced Needle, a 26M parameter function-calling (tool use) model. It runs at 6000 tok/s prefill and 1200 tok/s decode on consumer devices. We were always frustrated by the little effort made towards building agentic models that run on budget phones, so we conducted investigations that led to an observation: agentic experiences are built upon tool calling, and massive models are overkill for it. Tool calling is fundamentally retrieval-and-assembly (match query to tool name, extract argument values, emit JSON), not reasoning. Cross-attention…
Agentic Video Understanding in Gemini4d ago · producthunt.com · ▲147Agentic video analysis for faster, smarter Gemini insights Discussion | Link
ChatCut Desktop11d ago · chatcut.io · ▲290Video editor built for humans & AI. Edit with GPT & Claude.
More dev tools this month
the category →
The first open-source price index for GPU compute Discussion | Link
Dev tools · 10d ago · producthunt.com
Open-source GTM skills for technical founders
Dev tools · 29d ago · gtmcofounder.com
OpenTrailPaper is open-source bike computer firmware for the LilyGO T5S3 4.7" E-Paper PRO. It supports offline maps, GPX routes, FIT recording and Bluetooth sensors.
Dev tools · 2d ago · opentrailpaper.com
Launched alongside, March 2026
the whole month →
Switch from ChatGPT to Claude with import memory feature
AI · Mar 2026 · claude.com
