GEMINI LABJP
2.5ONLY — From the September 18 changelog: access to the 2.5 models is now limited to people who have actively used them. They are not deprecated and the API keeps serving them, while new projects are pointed at 3.5 Flash-Lite or 3.8 Flash3.8LIVE — Gemini 3.8 Live and 3.8 Live Extended Thinking went GA on September 15. Both are audio-to-audio models for the Live API, and the second keeps reasoning in the background during the call09/30 — gemini-omni-flash-preview shuts down on September 30, seven days away. Its successor, gemini-omni-1.1-flash, went GA on August 27ONCE — A report that "allow for this session" only lasts one call when the command contains a path has drawn 159 comments. The question underneath is how approvals should be scoped at allNEW — When a function call comes back as plain text, suspect the property names in your tool declaration firstHISTORY — Writing just three things outside the tool — what you settled on, the instructions you used, and why you redid something — keeps past work from disappearing with a model2.5ONLY — From the September 18 changelog: access to the 2.5 models is now limited to people who have actively used them. They are not deprecated and the API keeps serving them, while new projects are pointed at 3.5 Flash-Lite or 3.8 Flash3.8LIVE — Gemini 3.8 Live and 3.8 Live Extended Thinking went GA on September 15. Both are audio-to-audio models for the Live API, and the second keeps reasoning in the background during the call09/30 — gemini-omni-flash-preview shuts down on September 30, seven days away. Its successor, gemini-omni-1.1-flash, went GA on August 27ONCE — A report that "allow for this session" only lasts one call when the command contains a path has drawn 159 comments. The question underneath is how approvals should be scoped at allNEW — When a function call comes back as plain text, suspect the property names in your tool declaration firstHISTORY — Writing just three things outside the tool — what you settled on, the instructions you used, and why you redid something — keeps past work from disappearing with a model
TAG

Gemma

3 articles
Back to all tags
Related:
Gemma 42Ollama2local LLM2Gemini2API1premium1Windows1gemma-41open-source1edge-ai1moe1multimodal1
Gemini Dev/2026-06-24Intermediate

Folding a Local Gemma 4 into Daily Work — Practical Notes on the Ollama API and Response Speed

Taking a local Gemma 4 you can now run interactively and folding it into real work: how to hit Ollama's local API from a script, tricks to improve perceived response speed, and a two-tier fallback that automatically routes to the cloud Gemini API — code included.

Gemini Dev/2026-06-24Beginner

Running Gemma 4 Locally on Windows — A Hands-On LLM in Two Commands with Ollama

How to run Google's lightweight open model Gemma 4 locally on a Windows laptop. With Ollama, you go from install to running in effectively two commands. Plus how to split work between the cloud Gemini API and a local Gemma.

Gemini Advanced/2026-04-09Advanced

Gemma 4: From Edge E2B to Cloud 31B—Choosing the Right Model and Implementation Patterns

Comprehensive exploration of Google DeepMind's Gemma 4 family (E2B/E4B/26B A4B/31B). Master MoE architecture, 256K context windows, native thinking mode, and multimodal capabilities. Learn edge deployment strategies, production implementations, and fine-tuning best practices.