Gemini 3.7 Flash: Google’s Workhorse Coding Model at Half the Price, Three Weeks After 3.6

Google’s Gemini 3.7 Flash costs $0.75 per million input tokens, half of 3.6 Flash, while lifting its FrontierCode 1.1 coding score to 43.6% and DeepSWE v1.1 to 65.3%.
artificial-intelligence
Author

Kabui, Charles

Published

2026-08-20

Keywords

gemini-3-7-flash, google-deepmind, gemini-api, frontiercode, agentic-ai

Download as PDF

Google released Gemini 3.7 Flash on August 13, its most capable workhorse model for coding and agents, three weeks after Gemini 3.6 Flash. A workhorse model is the fast, cheaper option for high-volume everyday work, while bigger models handle the hardest problems. 3.7 Flash is built on 3.6 Flash with stronger reasoning, and ships at an introductory price of $0.75 per million input tokens and $3.75 per million output tokens, half of 3.6 Flash. On FrontierCode 1.1 it scores 43.6% versus 34.4%, and on DeepSWE v1.1 it reaches 65.3% versus 49.0%. Document reasoning jumps on GDP.pdf (34.0% versus 22.0%) and business workflow automation on AutomationBench (30.4% versus 17.0%). It keeps a 1 million token context window.

The practical win is price. A team running agents on 3.6 Flash gets better coding and tool use on 3.7 Flash at roughly half the input cost, important when agents send many small requests. The introductory rate runs through December 31, 2026, then rises to $1.50 per million input tokens, so the savings are time-limited. Google is also moving Gemini Spark, its 24/7 personal agent for AI Pro and Ultra subscribers, onto 3.7 Flash, making it more accurate on multi-step tasks in Google Workspace apps.

The release is also a statement about pace. Google has shipped a meaningfully better workhorse model three weeks after the previous one, and the biggest gains landed in the cheap tier, not the flagship. For everyday agent work, the frontier is moving fastest at the bottom of the price curve.

Read More: Gemini 3.1 Flash-Lite: Google’s Fastest Model Costs $0.25 per Million Tokens

Sources:

Disclaimer: For information only. Accuracy or completeness not guaranteed. Illegal use prohibited. Not professional advice or solicitation. Read more: /terms-of-service

Reuse

GNU GENERAL PUBLIC LICENSE v3.0(View License)

Citation

BibTeX citation:
@misc{kabui2026,
  author = {{Kabui, Charles}},
  title = {Gemini 3.7 {Flash:} {Google’s} {Workhorse} {Coding} {Model}
    at {Half} the {Price,} {Three} {Weeks} {After} 3.6},
  date = {2026-08-20},
  url = {https://toknow.ai/posts/gemini-3-7-flash-half-price-coding-workhorse/},
  langid = {en-GB}
}
For attribution, please cite this work as:
Kabui, Charles. 2026. “Gemini 3.7 Flash: Google’s Workhorse Coding Model at Half the Price, Three Weeks After 3.6.” https://toknow.ai/posts/gemini-3-7-flash-half-price-coding-workhorse/.