Gemini-3-flash-preview

Gemini 3 Flash Preview 是一款高速、高价值的思维模型，专为代理工作流程、多轮对话和编码辅助而设计。它提供接近 Pro 级别的推理和工具使用性能，同时延迟远低于大型 Gemini 变体，使其非常适合交互式开发、长时间运行的代理循环和协作编码任务。与 Gemini 2.5 Flash 相比，它在推理、多模态理解和可靠性方面提供了全面的质量改进。

能力

"POST", "endpoint": "/v1/chat/completions", "protocol": "openai-chat", "parameters": {"optional": {"stream": {"type": "boolean", "default": false}, "max_tokens": {"type": "integer", "default": 1000}, "temperature": {"max": 2, "min": 0, "type": "number", "default": 0.7}}, "required": ["model", "messages"]}, "validation": {"timeout_ms": 30000, "status_codes": [200], "response_path": "choices[0].message.content", "must_be_non_empty": true}, "test_request": {"model": "{{model}}", "messages": [{"role": "user", "content": "Say hello in one sentence"}], "max_tokens": 50, "temperature": 0.7}}

输入

text

输出

text

区域

global

验证时间

2026-08-15T04:50:43.133959782+00:00

价格

3.6 IFT / 1M tokens

文档

-

调用示例

curl https://api.inferlink.ai/v1/chat/completions \
  -H 'Authorization: Bearer $INFERLINK_API_KEY' \
  -H 'Content-Type: application/json' \
  -d '{"model":"google/google/gemini-3-flash-preview","messages":[{"role":"user","content":"Hello"}]}'