About Gemini 3.7 Flash
gemini-3.7-flash is a Multimodal LLM. It is Google's workhorse model for coding and agentic tasks, with substantial gains over Gemini 3.6 Flash in software engineering, web development, and dense-document comprehension (65.3% vs 49.0% on DeepSWE v1.1, 43.6% vs 34.4% on FrontierCode 1.1, 34.0% vs 22.0% on GDP.pdf).
Some other noteworthy features of gemini-3.7-flash include configurable thinking levels (low, medium, high; unlike earlier Flash models, minimal is not supported), structured output, tool use, code execution, context caching, search grounding, and support for multimodal inputs like text, images, audio, video, and PDFs. Multimodal gains over 3.6 Flash are uneven: long-video understanding improves modestly (85.4% vs 84.2% on LVBench) and chart reasoning is slightly lower on CharXiv, so document-heavy work benefits more than chart analysis.
It also supports Google's agentic video understanding, an opt-in per-request mode where the model decides which video segments to load, at what speed, and through which modality (frames, audio, or transcript) instead of sampling at a fixed frame rate. Google reports up to 88% fewer tokens, up to 66% lower cost, and up to 7% better accuracy on video analysis, at standard token pricing.
| Metric | Value |
|---|---|
| Parameter Count | Unknown |
| Mixture of Experts | Unknown |
| Context Length | 1M tokens |
| Multilingual | Yes |
| Quantized* | Unknown |
*Quantization is specific to the inference provider and the model may be offered with different quantization levels by other providers.
Ready to build with Gemini 3.7 Flash?
Try Gemini 3.7 Flash in the Workbench to prompt it, compare outputs, and iterate on prompts without writing any code. When you're ready to ship, call the same model from our API and build your own apps on top of it.
curl -sSf -X POST https://hub.oxen.ai/api/ai/chat/completions \
-H "Content-Type: application/json" \
-H "Authorization: Bearer $OXEN_API_KEY" \
-d '{
"model": "gemini-3-7-flash",
"messages": [
{
"role": "user",
"content": "Try sending a message."
}
]
}'