◉GEMINI LABJP
●CHAT — From August 26, Google Chat becomes the Ask Gemini hub for searching, drafting, catching up on threads, and managing tasks and events with Workspace context intact. Two days out●ANDROID — Gemini replaces Google Assistant on Android from September 4, eleven days from now. Now is the time to check any voice shortcuts you built on Assistant●ROBOTICS — gemini-robotics-er-1.6-preview shuts down on August 31, seven days out. The ER 2 line succeeds it with spatial reasoning, multi-step tool orchestration, and multi-robot coordination●PRICE — Gemini 3.7 Flash introductory pricing is $0.75 input and $3.75 output per million tokens through December 31. From January 1, 2027 it doubles to $1.50 and $7.50●FREE — Google AI Studio still offers a free API tier with daily request limits and no credit card. If you only want to see how 3.7 Flash behaves, that is enough to start●SCALE — The Gemini app crossed one billion monthly users on August 11. The split is settling in: 3.1 Pro for deep reasoning, the Flash line for production work where speed and unit cost decide●CHAT — From August 26, Google Chat becomes the Ask Gemini hub for searching, drafting, catching up on threads, and managing tasks and events with Workspace context intact. Two days out●ANDROID — Gemini replaces Google Assistant on Android from September 4, eleven days from now. Now is the time to check any voice shortcuts you built on Assistant●ROBOTICS — gemini-robotics-er-1.6-preview shuts down on August 31, seven days out. The ER 2 line succeeds it with spatial reasoning, multi-step tool orchestration, and multi-robot coordination●PRICE — Gemini 3.7 Flash introductory pricing is $0.75 input and $3.75 output per million tokens through December 31. From January 1, 2027 it doubles to $1.50 and $7.50●FREE — Google AI Studio still offers a free API tier with daily request limits and no credit card. If you only want to see how 3.7 Flash behaves, that is enough to start●SCALE — The Gemini app crossed one billion monthly users on August 11. The split is settling in: 3.1 Pro for deep reasoning, the Flash line for production work where speed and unit cost decide
TAG

Long-Term Memory

1 articles
← Back to all tags
Related:
Google ADK1SessionService1Conversation Compaction1Gemini1Python1Eval Harness1
⬡ Gemini Advanced/2026-07-10Advanced

My ADK Assistant Quietly Forgot a Deadline — Catching Compaction Memory Loss With a Recall Probe

Compacting conversation history in Google ADK with Gemini lowers cost, but it also erodes what your assistant remembers — silently. Here is how I built a recall probe to measure that loss, compared three compaction strategies against the same ledger, and stopped trading memory for tokens.