DeepSeek V3.1
by DeepSeekRank #152Score 71.8
DeepSeek-V3.1 is a large hybrid reasoning model (671B parameters, 37B active) that supports both thinking and non-thinking modes via prompt templates. It extends the DeepSeek-V3 base with a two-phase long-context...
性能概览
评分
71.8
排名
#152
24小时变化
+1
7天变化
+7
状态
stable置信度
high信号评分
| 信号 | 标准化 | 权重 | 贡献 | 新鲜度 |
|---|---|---|---|---|
Capabilities capability | 66.7 | 20% | 13.3 | 2026-09-22T15:11:14.936Z |
Benchmarks benchmark | 68.9 | 30% | 20.7 | 2026-09-22T15:11:14.936Z |
Pricing pricing_tier | 99.1 | 15% | 14.9 | 2026-09-22T15:11:14.936Z |
Context Window context_window | 82.8 | 10% | 8.3 | 2026-09-22T15:11:14.936Z |
Recency recency | 60.5 | 15% | 9.1 | 2026-09-22T15:11:14.936Z |
Output Capacity output_capacity | 72.2 | 10% | 7.2 | 2026-09-22T15:11:14.936Z |
主要驱动因素
positive
Benchmarks
positive
Pricing
$0.95/M output tokens
$0.95
positive
Capabilities
Supports reasoning, tools, JSON mode, streaming
4/7
positive
Recency
Released over 1 year ago
1.1y ago
功能
| 功能 | 支持 |
|---|---|
| 视觉 | 否 |
| 推理 | 是 |
| JSON模式 | 是 |
| 流式输出 | 是 |
| 函数调用 | 是 |
| 网页搜索 | 否 |
价格
输入/百万令牌
$0.25
输出/百万令牌
$0.95
上下文窗口
164K
最大输出
33K