Key Highlights
- Google unveiled three Gemini AI models: Gemini 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber
- The 3.6 Flash model achieves 17% lower output token consumption compared to previous versions, with pricing set at $1.50 per million input tokens
- A specialized security model, Gemini 3.5 Flash Cyber, will be accessible exclusively to government entities and vetted partners through CodeMender
- The lightweight 3.5 Flash-Lite delivers 350 tokens per second output speed, available at $0.30 per million input tokens
- These releases precede Alphabet’s upcoming quarterly financial report, with GOOGL shares declining 0.66% to $349.66
On Tuesday, Alphabet introduced a trio of Gemini AI models, timing the announcement mere days before its scheduled quarterly financial disclosure. Trading activity showed GOOGL shares at $349.66, representing a 0.66% decline.
The newly introduced models ā Gemini 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber ā target developers and enterprise clients constructing AI applications requiring substantial scale.
As the flagship offering, Gemini 3.6 Flash demonstrates superior capabilities over 3.5 Flash across coding, knowledge processing, and multimodal operations, while achieving 17% reduction in output token consumption. Its pricing structure stands at $1.50 per million input tokens and $7.50 per million output tokens.
Benchmark testing reveals 3.6 Flash achieves 49% on DeepSWE, advancing from the 37% recorded by 3.5 Flash. MLE Bench results show improvement to 63.9% from 49.7%. Performance on OSWorld-Verified for computer use tasks climbed to 83.0% from 78.4%.
Early adopters including Figma, Harvey, Hebbia, and JetBrains have integrated the model. Hebbia and Harvey specifically highlighted its multimodal capabilities ā especially document processing and visual chart interpretation.
High-Speed Performance with Gemini 3.5 Flash-Lite
Designed for high-volume operations requiring rapid processing, the 3.5 Flash-Lite model delivers 350 output tokens per second at a cost structure of $0.30 per million input tokens and $2.50 per million output tokens.
Flash-Lite demonstrates substantial performance gains over 3.1 Flash-Lite in agentic operations and coding challenges. Terminal-Bench 2.1 results show 54% versus 31%. SWE-Bench Pro testing yields 54.2% compared to 49.6% for the earlier 3 Flash version.
Early enterprise clients including Palo Alto Networks and Ramp have emphasized the model’s velocity and economic efficiency for expanding operational workflows.
Security-Specialized Model Enters the Arena
Gemini 3.5 Flash Cyber represents a dedicated security solution, refined from 3.5 Flash specifically for identifying and resolving software vulnerabilities. It functions within Google‘s CodeMender agent system, where multiple Flash Cyber agents collaborate to generate comprehensive security assessments.
Google has adopted a restricted distribution strategy for this technology. Access remains limited to government agencies and approved partners through a controlled pilot program, reflecting concerns about potential dual-use applications.
CyberGym benchmark results show Flash Cyber achieving what Google characterizes as competitive frontier-level performance.
This cybersecurity initiative positions Google in heightened competition with OpenAI and Anthropic, both having recently introduced capabilities targeting security professionals and vulnerability analysis.
Google’s premium Gemini 3.5 Pro model remains under evaluation with select partners. Initially anticipated for June release, no definitive launch timeline has been announced. The company has verified that pre-training for Gemini 4 is underway.
Both 3.6 Flash and 3.5 Flash-Lite have achieved general availability through Google AI Studio, Android Studio, Gemini Enterprise, and the Gemini application. Flash-Lite is additionally being deployed within Google Search infrastructure.
Get 3 Free Stock Ebooks
Discover top-performing stocks in AI, Crypto, and Technology with expert analysis.
- Top 10 AI Stocks - Leading AI companies
- Top 10 Crypto Stocks - Blockchain leaders
- Top 10 Tech Stocks - Tech giants
