Gemini 3.5 Flash + Pro: Powerful, Cheap, & Fast NEW AI Model! (Fully Tested)

2026-05-16 · Source: WorldofAI · Field: Technology & Digital — Artificial Intelligence & Machine Learning · Depth: Intermediate, medium

Summary

Google is preparing to launch Gemini 3.5 Flash and Pro variants at the upcoming Google I/O conference. Initial internal testing of the Gemini 3.2 Pro variant, codenamed Cappuccino, showed underwhelming results due to repetitive UI generation. However, the newly renamed Gemini 3.5 Flash checkpoint, despite its "flash tier" designation, demonstrates remarkable output quality, including generating a functional Minecraft clone with background music and server connectivity. This Flash variant also excels in ASCII art generation, showcasing strong spatial consistency and token-level planning. While the Gemini 3.5 Flash is fast and intelligent with improved instruction following and less repetitive UI, it still suffers from over-generation of excessive UI elements and occasional prompt adherence issues, such as ignoring instructions to avoid web access during testing. The Mac OS clone generation was also noted as less impressive compared to a previous Gemini 3.2 Pro variant.

Key takeaway

For AI engineers evaluating new large language models for code generation, you should prioritize testing Gemini 3.5 Flash for its demonstrated ability to create complex, interactive applications like a Minecraft clone. Be aware of its tendency to generate excessive UI elements and verify prompt adherence, particularly regarding web access, to ensure it meets your specific control requirements.

Key insights

Gemini 3.5 Flash and Pro models show significant qualitative improvements, especially in complex code generation.

Principles

• Flash tier models can achieve high-quality code generation.
• Spatial consistency is critical for advanced text-based outputs.

Method

Internal testing involved evaluating Gemini 3.2 and 3.5 checkpoints (e.g., Fanta, Sprite, Cola, Cappuccino) for code generation quality and UI consistency.

In practice

• Explore Gemini 3.5 Flash for complex UI and game code generation.
• Test prompt adherence carefully, especially for web access instructions.

Topics

• Gemini 3.5 Flash
• Gemini 3.5 Pro
• Code Generation
• UI Generation
• Spatial Consistency

Best for: AI Engineer, CTO, VP of Engineering/Data, AI Scientist, Machine Learning Engineer, Tech Journalist

Related on AIssential

• Gemini 3.5 Flash: Google's Most Powerful Model Ever! Beats Opus 4.7 & GPT 5.5? (Fully Tested) — Gemini 3.5 Flash offers superior coding and creative generation at a higher cost and token usage than previous flash models.
• Gemini 3.5 Flash has landed. — Gemini 3.5 Flash offers frontier intelligence with four times faster output speed and improved benchmark performance.
• Gemini 3.0 Pro GA WILL BE Google's Greatest Model Ever! Most Powerful AI EVER! (Early Test) — Gemini 3.0 GA shows strong agentic task performance and advanced code generation, particularly for SVG and functional UI.
• Gemini 3.5 Flash Test | Coding, OCR, Image Understanding, Pricing, Speed | 🔴 Live — Gemini 3.5 Flash offers significant speed improvements and strong document understanding, but its coding and frontend capabilities are inconsistent.
• How to Use Gemini 3 Pro in CLI? — Gemini 3 Pro in CLI enhances developer workflows through advanced AI reasoning, multimodal capabilities, and long-context processing.
• Gemini 3.5 Flash might be fast enough for gen AI to make sense — Gemini 3.5 Flash demonstrates that high intelligence and efficiency can enable scalable agentic AI.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by WorldofAI.