Google DeepMind's Gemini 3.7 Flash (Aug 13, 2026) is the most intelligent workhorse model in the Gemini line for coding and agents. It cuts the introductory price 50% versus 3.6 Flash to $0.75 / $3.75 per 1M tokens (held through Dec 31, 2026) while lifting coding benchmarks: FrontierCode 43.6%, DeepSWE 65.3%, WebDev Arena Elo 1588, AutomationBench 30.4%. It runs on a 1M-token multimodal context across Google AI Studio and the Gemini Enterprise Agent Platform in 160+ countries.
Parameters
Undisclosed
Context Window
1M
License
Proprietary
Release Date
2026-08-13
API Pricing
Input Price (per 1M tokens)
$0.75
Output Price (per 1M tokens)
$3.75
Billing Mode: standard
Strengths
- •Best-in-class Flash coding & agent performance
- •50% lower price than 3.6 Flash
- •1M-token multimodal context
- •Available in 160+ countries
Weaknesses
- •Positioned as workhorse, not the top reasoning tier
- •Trails Pro-tier on hardest reasoning
- •Introductory price is time-limited (through Dec 31, 2026)
Use Cases
- •Coding assistants & software agents
- •Web and app development
- •Multimodal input processing
- •Cost-sensitive production workloads