Google Surprise-Drops Gemini 3.6 Flash with Cheaper Pricing for Agentic Tasks
AGI Hunt · wechat · 2026-07-21
Without official promotion, Google has quietly added two new models to AI Studio: Gemini 3.6 Flash and Gemini 3.5 Flash Lite.
- Gemini 3.6 Flash: Described as "our most intelligent model yet," it focuses on delivering consistent frontier performance in agentic and coding tasks. Priced at $1.50 input / $7.50 output per million tokens, it actually lowers the output cost compared to the existing Gemini 3.5 Flash ($9.00 output) while improving performance.
- Gemini 3.5 Flash Lite: Designed for large-scale, high-frequency agentic tasks and subagent workflows, emphasizing high throughput and low latency. Priced at just $0.30 input / $2.50 output, it is ideal for the heavy lifting of parallel tasks delegated by a main agent.
Additionally, the author noted conflicting descriptions for 3.6 Flash across different pages and a somewhat confusing version naming scheme. The Pro version is notably absent, while the two new models are already available for testing.
Related event: Google Launches Three New Gemini Models, Announces Gemini 4 Pre-training(157 posts)→
More from Models
- Daily AI brief: GPT-Live-1 in API, OpenAI pauses $200 Pro signups amid Astra demand — koltregaskes · 2026-09-11
- Same Echo Maze prompt, three frontier models: all passed visually but shipped the same hidden bug — eyishazyer · 2026-09-11
- Benchmark scores drop from 89% to 19% on new evals — how benchmaxxing breaks leaderboard trust — airesearch12 · 2026-09-11
- ChatGPT tells user their question is too hard and to 'accept dumber answers' — phido3000 · 2026-09-11
- Developer Building a Unified Leaderboard of All Model Benchmark Scores — airesearch12 · 2026-09-11
- Rumor claims Kimi faked performance by serving Claude; DeepSeek new model surprises in evals — realsohamparekh · 2026-09-11