utWhat it does
Published capability labels: Vision, Video understanding, Audio understanding, Reasoning, Tool use. Listed uses: Chat & writing, Reasoning, Tool use.
- Inputs
- text, image, video, audio, pdf
- Outputs
- text
- Category
- Language
- Context window
- 1.05M tokens
- License
- Provider terms
These labels describe intended capabilities. Scout has not independently evaluated this model or established that it is best at a particular task.
Cloud pricing
Listed base text API rates from Models.dev, in USD per million tokens. These are not consumer chat subscription prices.
| Offering | Input | Output | Cached input |
|---|---|---|---|
| Google · gemini-3.7-flash ↗ | $0.75 | $3.75 | $0.075 |
Rates may change; higher context or usage tiers can cost more. Check the provider for current availability and tool, media, tax, or other charges.
Tested performance
Independent leaderboard results for exact configurations of this model. Each test has its own scale, so compare results within a test, not across tests. See every tested model.
Code Arena
Build apps & websites · Ratings published Oct 23, 2026
| Configuration | Result | 95% interval | Evidence |
|---|---|---|---|
| Gemini 3.7 Flash (High) | 1590.3 | 1583.6 to 1597.0 | 13,307 votes · leaderboard rank #33 |
A preference rating, not a percentage of correct answers. Design, usability, and the coding setup affect votes. Original leaderboard ↗ · CC BY 4.0
Terminal-Bench 4.0
Complete terminal tasks · Latest imported submission Oct 6, 2026
| Configuration | Result | 95% interval | Evidence |
|---|---|---|---|
| gemini/gemini-3.7-flash | 11.21% | 8.76% to 13.66% | 330 trials · leaderboard rank #19 |
Agent versions and spending differ. These results do not isolate model ability or predict speed on your computer. Original leaderboard ↗ · Apache 2.0
Agent Arena
Work with an AI agent · Ratings published Oct 23, 2026
| Configuration | Result | 95% interval | Evidence |
|---|---|---|---|
| Gemini 3.7 Flash (High) | -0.013 | -0.022 to -0.003 | 56,113 sessions · leaderboard rank #29 |
A net improvement index relative to Arena’s baseline. Zero is the baseline, not zero capability; this is not a task completion percentage. Original leaderboard ↗ · CC BY 4.0
Benchmark evidence
No vendor-reported results are attached to this listing.