Gemini 3.7 Flash vs Gemini 3.6 Flash
Gemini 3.7 Flash landed August 13, 2026, three weeks after Gemini 3.6 Flash, and unlike that release this one moves capability as well as cost. Google says it did not retrain from scratch: 3.7 Flash is algorithmic improvements plus user feedback applied to the 3.6 line, shipped as a full replacement rather than a parallel option. Pricing is the loudest number. Introductory rates run $0.75 per million input tokens and $3.75 per million output through December 31, 2026, exactly half of 3.6 Flash, then revert to $1.50/$7.50, which is to say back to parity. The coding gains Google publishes are larger than the 3.6 step was: DeepSWE v1.1 from 48.6 to 65.3 percent, FrontierCode 1.1 Main from 34.4 to 43.6, Terminal-bench 2.1 at 85.8, and GDM-MRCR v2 at 97.0 percent for 128K retrieval. Two caveats keep this honest. First, every figure here is Google-reported; Artificial Analysis has no SWE-bench Verified score for 3.7 Flash yet, and 3.6 Flash is the cautionary tale, since its Intelligence Index came back flat against 3.5 despite a strong vendor table. Second, the price is promotional. If you build a unit-economics model on $0.75/$3.75, put a January 1, 2027 reminder in it. For anything currently on 3.6 Flash, the migration case is straightforward through the end of the year: same window, same modalities, lower rate, better published coding scores.
Head-to-Head Specs
| Spec | Gemini 3.7 Flash | Gemini 3.6 Flash |
|---|---|---|
| Provider | ||
| Input Price | $0.75/1M | $1.50/1M |
| Output Price | $3.75/1M | $7.50/1M |
| Context Window | 1.0M | 1.0M |
| Released | 2026-08 | 2026-07 |
| Capabilities | text, vision, audio, video, tool-use, code, reasoning | text, vision, audio, video, tool-use, code, reasoning |
Category Breakdown
$0.75/$3.75 introductory against $1.50/$7.50, exactly half on both sides of the meter
3.7 Flash reverts to $1.50/$7.50, identical to 3.6 Flash, so the advantage is time-limited by design
Google reports 65.3 percent against 48.6, a 16.7 point step and much larger than the 3.5 to 3.6 move
Google reports 43.6 percent against 34.4
97.0 percent on GDM-MRCR v2 at 128K, with the same 1,048,576 token input window on both
Both take 1,048,576 input tokens, up to 65,536 output, and text, image, audio, video, and PDF input
3.6 Flash has Artificial Analysis measurements on record; 3.7 Flash has no independent SWE-bench Verified score yet
3.6 Flash has three weeks of settled prompt behavior, which is thin but is more than zero
Choose Gemini 3.7 Flash when:
- ▸Any 3.6 Flash workload that can re-run its evals before January 2027
- ▸Agentic coding and web development where the DeepSWE gap shows up in fewer retries
- ▸Long-context retrieval over large document sets
- ▸High-volume agent pipelines where a 50 percent rate cut compounds fast
Choose Gemini 3.6 Flash when:
- ▸Pipelines with prompts and token budgets tuned tightly to 3.6 Flash verbosity
- ▸Procurement that needs a rate valid past December 31, 2026 without a re-quote
- ▸Teams that will only move on independently verified benchmarks
- ▸Evaluation harnesses mid-run that a model swap would invalidate
Frequently Asked Questions
Which is better, Gemini 3.7 Flash or Gemini 3.6 Flash?
It depends on your use case. Gemini 3.7 Flash from Google excels at any 3.6 flash workload that can re-run its evals before january 2027, while Gemini 3.6 Flash from Google is better for pipelines with prompts and token budgets tuned tightly to 3.6 flash verbosity. See the full comparison above for detailed benchmarks and pricing.
How much does Gemini 3.7 Flash cost compared to Gemini 3.6 Flash?
Gemini 3.7 Flash costs $0.75 input and $3.75 output per 1M tokens. Gemini 3.6 Flash costs $1.50 input and $7.50 output per 1M tokens.
What is the context window difference between Gemini 3.7 Flash and Gemini 3.6 Flash?
Gemini 3.7 Flash supports 1.0M tokens, while Gemini 3.6 Flash supports 1.0M tokens.