Model
Gemini 3 Flash Preview
New Gemini flash lane bringing frontier-style multimodal reasoning to cheaper runs
11 providersreleased knowledge cutoff 2025-01closed weights
$0.5
best input /M · OrcaRouter
$3
best output /M
1.05M
context window
66K
max output
Compare providers (sorted by listed input price)
| Provider | Input /M | Output /M | Cache read /M | Cache write /M | Context | Capabilities | Status | Updated |
|---|---|---|---|---|---|---|---|---|
| OrcaRouter google/gemini-3-flash-preview | $0.5 | $3 | $0.05 | — | 1.05M | reasoningtoolsstructuredvision | Dec 17, 2025 | |
| NanoGPT google/gemini-3-flash-preview | $0.5 | $3 | $0.05 | — | 1.05M | reasoningtoolsstructuredvision | Dec 17, 2025 | |
| Google gemini-3-flash-preview | $0.5 | $3 | $0.05 | — | 1.05M | reasoningtoolsstructuredvision | Dec 17, 2025 | |
| OpenRouter google/gemini-3-flash-preview | $0.5 | $3 | $0.05 | $0.083 | 1.05M | reasoningtoolsstructuredvision | Dec 17, 2025 | |
| Merge Gateway google/gemini-3-flash-preview | $0.5 | $3 | $0.05 | — | 1.05M | reasoningtoolsstructuredvision | Dec 17, 2025 | |
| ZenMux google/gemini-3-flash-preview | $0.5 | $3 | $0.05 | $1 | 1.05M | reasoningtoolsno structuredvision | Dec 17, 2025 | |
| Perplexity Agent google/gemini-3-flash-preview | $0.5 | $3 | $0.05 | — | 1.05M | reasoningtoolsno structuredvision | Dec 17, 2025 | |
| Ofox google/gemini-3-flash-preview | $0.5 | $3 | $0.05 | $1 | 1.05M | reasoningtoolsstructuredvision | Dec 17, 2025 | |
| Eden AI google/gemini-3-flash-preview | $0.5 | $3 | $0.05 | $0.083 | 1.05M | reasoningtoolsstructuredvision | Dec 17, 2025 | |
| Kilo Gateway google/gemini-3-flash-preview | $0.5 | $3 | $0.05 | $0.083 | 1.05M | reasoningtoolsstructuredvision | Dec 17, 2025 | |
| AnyAPI google/gemini-3-flash-preview | — | — | — | — | 1.05M | reasoningtoolsstructuredvision | Dec 17, 2025 |
Price badge
https://models.sutraworks.ai/badge/google/gemini-3-flash-preview.svgLive SVG, regenerated on every hourly sync — paste it into a README to always show the current cheapest listed price. Updates when prices change; no build step needed on your side.
Benchmarks
- SWE-Bench Pro ↗(resolve rate)report ↗34.63
- SWE-Atlas Codebase QnA(score)report ↗8.2
- SWE-Atlas Refactoring(score)report ↗10
- SWE-Atlas Test Writing(score)report ↗30.3
Prices are per million tokens (USD). “—” means the provider does not publicly list a price for this model. Data from models.dev; verify with the provider before purchasing.