TLDR
- Google unveiled a trio of Gemini AI models: 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber
- The new 3.6 Flash model delivers 17% lower output token consumption compared to the previous version, with pricing at $1.50 per million input tokens
- Flash Cyber represents a specialized security model restricted to government entities and select partners through CodeMender
- Flash-Lite achieves 350 tokens per second output speed 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
Google introduced three advanced Gemini AI models this Tuesday, timing the announcement just before Alphabet prepares to unveil its quarterly financial results. Shares of GOOGL were changing hands at $349.66, representing a 0.66% decline during trading.
The trio of releases ā consisting of Gemini 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber ā targets developers and enterprise clients seeking to deploy AI-driven solutions at significant scale.
Taking center stage is Gemini 3.6 Flash. According to Google, this model surpasses the performance of 3.5 Flash across coding tasks, knowledge-based operations, and multimodal applications, all while consuming 17% fewer output tokens. The pricing structure sits at $1.50 for each million input tokens and $7.50 for every million output tokens.
When evaluated on coding benchmarks, 3.6 Flash achieves a 49% score on DeepSWE, representing an increase from the 37% recorded by 3.5 Flash. The model reaches 63.9% on MLE Bench, climbing from the previous 49.7%. Performance metrics for computer use on OSWorld-Verified also demonstrate improvement, rising to 83.0% from 78.4%.
Early adopters like Figma, Harvey, Hebbia, and JetBrains have already integrated the model into their operations. Both Hebbia and Harvey highlighted its robust multimodal capabilities, especially for document processing and chart interpretation.
Gemini 3.5 Flash-Lite: Velocity Meets Affordability
Designed specifically for high-volume operations where processing speed is critical, the 3.5 Flash-Lite model delivers 350 output tokens per second. Its cost structure is set at $0.30 per million input tokens and $2.50 per million output tokens.
Flash-Lite demonstrates substantial performance gains over its predecessor, 3.1 Flash-Lite, particularly in agentic operations and coding challenges. The model achieves a 54% score on Terminal-Bench 2.1 compared to 31% previously. On SWE-Bench Pro, it reaches 54.2%, outperforming the earlier 3 Flash model’s 49.6%.
Palo Alto Networks and Ramp number among the initial adopters, emphasizing its velocity and economic efficiency for expanding operational workflows.
Cybersecurity Gets Its Own Model
Gemini 3.5 Flash Cyber represents a dedicated security solution, fine-tuned from the 3.5 Flash foundation to identify and remediate software vulnerabilities. The model functions within Google’s CodeMender agent infrastructure, where multiple Flash Cyber agents collaborate to generate comprehensive security assessments.
Google is implementing a restricted rollout strategy for this technology. Access to the model will be limited exclusively to government organizations and vetted partners through a controlled pilot program, acknowledging the dual-use considerations associated with the technology.
When tested on the CyberGym benchmark, Flash Cyber demonstrates what Google characterizes as competitive frontier-level performance.
This cybersecurity initiative positions Google more directly against competitors like OpenAI and Anthropic, both of which have recently introduced solutions targeting security professionals and vulnerability identification.
Google’s premium Gemini 3.5 Pro model remains under evaluation with select partners. Initially anticipated for a June release, the launch timeline remains unconfirmed. The company has also acknowledged commencing pre-training activities for Gemini 4.
Both 3.6 Flash and 3.5 Flash-Lite are immediately accessible through Google AI Studio, Android Studio, Gemini Enterprise, and the Gemini application. Flash-Lite is additionally being deployed within Google Search.
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
