🤖Google Ships Gemini Omni, 3.5 Flash, Spark Agent at I/O
TL;DR
Google launched Gemini 3.5 Flash, a multimodal Omni model, and the Gemini Spark agent at I/O 2026, alongside a $50/month cut to the AI Ultra plan. Flash hits 4x the speed of comparable models at $1.50/$9 per 1M tokens with a 1M-token context window.
Google launched Gemini 3.5 Flash, a multimodal Omni model, and the Gemini Spark agent at I/O 2026, alongside a $50/month cut to the AI Ultra plan. Flash hits 4x the speed of comparable models at $1.50/$9 per 1M tokens with a 1M-token context window.
Key Points
Gemini 3.5 Flash: 1M-token context, $1.50/$9 per 1M tokens, 76.2% on Terminal-Bench 2.1
Gemini Omni accepts any input modality starting with video, rolling out via Gemini app and Flow
Gemini Spark is a general-purpose agent that reasons across connected apps, in beta for Ultra subscribers
AI Ultra subscription dropped from $250 to $200/month
Gemini 3.5 Pro stays internal, with broader release next month
Why It Matters
Google's playbook is now speed-and-price for Flash, agentic reach for Spark, and any-input coverage for Omni. The Ultra price cut signals the consumer LLM tier is becoming a margin business, not a moat.
Quick Facts
Frequently Asked Questions
Why does this matter?
Google's playbook is now speed-and-price for Flash, agentic reach for Spark, and any-input coverage for Omni. The Ultra price cut signals the consumer LLM tier is becoming a margin business, not a moat.
What happened?
Google launched Gemini 3.5 Flash, a multimodal Omni model, and the Gemini Spark agent at I/O 2026, alongside a $50/month cut to the AI Ultra plan. Flash hits 4x the speed of comparable models at $1.50/$9 per 1M tokens with a 1M-token context window.
Comments
Be the first to comment
Enjoyed this article?
Get it daily. 7am. Free. Reads in 5 minutes.
Join 3,131 builders reading daily.