Google's Gemini 3.6 Flash outscores 'Pro' version in new benchmarks

Google has updated its Gemini offerings, introducing Gemini 3.6 Flash and Gemini 3.5 Flash-Lite, while Gemini 3.1 Pro remains the default 'Pro' option in the Gemini application. Independent benchmarks from Artificial Analysis show Gemini 3.6 Flash scoring 50 on an intelligence index, surpassing the 'Pro' version (Gemini 3.1 Pro) which scored 46. Despite the 'Pro' designation, Gemini 3.1 Pro is an older model, and the more advanced Gemini 3.5 Pro is still in partner testing and has not yet been broadly released due to not meeting internal benchmarks for complex reasoning and coding. AI

IMPACT Confirms that performance gains in Flash models are outpacing 'Pro' versions, potentially shifting user perception and pricing strategies.

RANK_REASON Significant update to a major AI model family from a leading provider, with implications for pricing and performance perception. [lever_c_demoted from significant: ic=1 ai=1.0]

Read on dev.to — LLM tag →

• Artificial Analysis
• Bloomberg
• Copilot
• Gemini
• Gemini 3.1 Pro
• Gemini 3.5 Pro
• Gemini 3.6 Flash
• GitHub
• Google AI Pro
• Google DeepMind
• TechCrunch
• Tulsee Doshi

• model release
• product

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →