The latest preview version of Google's most advanced reasoning Gemini model, capable of solving complex problems. Built for the agentic era with enhanced reasoning capabilities, multimodal understanding (text, image, video, audio), and a 1M token context window. Features thinking preview, code execution, grounding with Google Search, system instructions, function calling, and controlled generation. Supports up to 3,000 images per prompt, 45-60 minutes of video, and 8.4 hours of audio.
发布日期2025年6月5日
参数规模—
上下文长度1.0M
许可证Proprietary
知识截止2025年1月31日
Benchmarks
评测成绩
| 评测基准 | 类别 | 分数 | 来源 |
|---|---|---|---|
| Global-MMLU-Lite | general language reasoning | 89.2 | 来源 |
| AIME 2025 | math reasoning | 88.0 | 来源 |
| FACTS Grounding | reasoning | 87.8 | 来源 |
| GPQA | reasoning general | 86.4 | 来源 |
| VideoMMMU | multimodal vision reasoning | 83.6 | 来源 |
| Aider-Polyglot | general code | 82.2 | 来源 |
| MMMU | multimodal reasoning general | 82.0 | 来源 |
| LiveCodeBench | reasoning general code | 69.0 | 来源 |
| SWE-Bench Verified | reasoning frontend_development code | 67.2 | 来源 |
| Vibe-Eval | multimodal vision general | 67.2 | 来源 |
| SimpleQA | general reasoning | 54.0 | 来源 |
| Humanity's Last Exam | general | 21.6 | 来源 |
| MRCR v2 (8-needle) | long_context reasoning general | 16.4 | 来源 |
Pricing
API 价格对比
| 服务商 | 输入价 | 输出价 | 上下文 | 吞吐(tok/s) | 延迟(s) | 函数调用 | 代码执行 | 联网搜索 |
|---|---|---|---|---|---|---|---|---|
| Jiekou.AI | $1.13 | $9.00 | 1.0M | — | — | ✓ | ✗ | ✗ |
| NanoGPT | $2.50 | $10.00 | 1.0M | — | — | ✗ | ✗ | ✗ |
价格单位：美元/百万 token，数据来自社区整理，仅供参考。