◉GEMINI LABJP
●CLI — Gemini CLI's nightly (Oct 2) makes chat recording append-only and bounds the history window. It should help with the weight of long sessions●12/31 — 89 days left until the introductory pricing ends. 3.8 / 3.7 / 3.6 Flash move to $1.50 input and $7.50 output, from $0.75 and $3.75●ACP — A report (#29595) says MCP tool permission requests do not show which server they come from. How to read the prompt is the open question●NEW — Docs, Sheets, Slides and Drive gained Gemini features in March. Here is what I kept and what I switched off six months later●WINDOWS — A report (#29614) says Windows pins GIT_CONFIG_GLOBAL=NUL and breaks Git. Worth checking if you run the CLI on Windows●LITE — gemini-3.1-flash-lite is scheduled to shut down on May 7, 2027, with gemini-3.5-flash-lite as the successor. No rush, but a good moment to draft a migration table●CLI — Gemini CLI's nightly (Oct 2) makes chat recording append-only and bounds the history window. It should help with the weight of long sessions●12/31 — 89 days left until the introductory pricing ends. 3.8 / 3.7 / 3.6 Flash move to $1.50 input and $7.50 output, from $0.75 and $3.75●ACP — A report (#29595) says MCP tool permission requests do not show which server they come from. How to read the prompt is the open question●NEW — Docs, Sheets, Slides and Drive gained Gemini features in March. Here is what I kept and what I switched off six months later●WINDOWS — A report (#29614) says Windows pins GIT_CONFIG_GLOBAL=NUL and breaks Git. Worth checking if you run the CLI on Windows●LITE — gemini-3.1-flash-lite is scheduled to shut down on May 7, 2027, with gemini-3.5-flash-lite as the successor. No rush, but a good moment to draft a migration table
TAG

video understanding

3 articles
← Back to all tags
Related:
Gemini API2multimodal2Files API2indie development1Gemini Omni Flash1cost design1indie developer1
⬡ Gemini Advanced/2026-09-05Advanced

Two Kinds of Video Questions: Why I Send "Find It" and "Prove It Isn't There" Down Separate Paths

Agentic video understanding lets the model decide which parts of a video to watch. That works beautifully for finding things, and it quietly breaks when you need to prove something never appears. Here is how I split my questions, and the coverage check I now run first.

◈ Gemini API/2026-07-05Advanced

Collapsing Video Understanding into One Native Call with Omni Flash

How I replaced an ffmpeg frame-extraction pipeline (7-9 calls per clip) with a single native Omni Flash call, the measured differences, and the boundaries where keeping frame sampling still wins.

◈ Gemini API/2026-06-28Advanced

Read Video with Timestamps in the Gemini API: Pull Just the Scene You Need

Hunting for 'where was that step?' in a screen recording or app demo is a chore. Here is how to use Gemini API video understanding to pull just the right scene with timestamps, plus a design that keeps tokens down with FPS and resolution.