See key specs and per-benchmark scores for each model/mode. Scroll horizontally for all columns. 当前对比 3 个模型的评测数据与核心参数。
DeepSeek-V3.1 并不是一次彻底的架构革新,而是对 V3 系列在 稳定性、推理性能与代码生成能力 上的平衡升级,同时在推理模式和 Agent 应用层面有了实质性进展。
与前代模型相比,V3.1 在 “thinking 模式” 与 “normal 模式” 上的差异化表现非常明显:
这意味着 V3.1 实现了推理与高效执行之间的动态平衡,适合在不同任务下灵活调用,而不是单一走“极致推理”路线。
V3.1 在 Agent 场景中的表现也有明显改进:
整体来看,V3.1 的核心价值在于:用混合模式和强化 Agent 能力,推动大模型从“只会答题”走向“能规划、能执行”的下一步。
DeepSeek-V3.1
DeepSeek-R1-0528
DeepSeek-V3-0324
Compare benchmark results across thinking modes and tool usage.
Data sourced primarily from official releases (GitHub, Hugging Face, papers), then benchmark leaderboards, then third-party evaluators. Learn about our data methodology
Best Overall
—
Best Single
—
Thinking Mode (Default)
DeepSeek-V3.1 · 1 Modality support
Higher is usually better; “—” means no score.
Complete scores for each model/mode across selected benchmarks.
Feature compare
Licensing, MoE architecture, and multi-modality support.
| Features & specs | DE DeepSeek-V3.1DeepSeek-AI | DE DeepSeek-R1-0528DeepSeek-AI | DE DeepSeek-V3-0324DeepSeek-AI |
|---|---|---|---|
Model snapshots | |||
Organization | DeepSeek-AI | DeepSeek-AI | DeepSeek-AI |
模型全名 | DeepSeek-V3.1 | DeepSeek-R1-0528 | DeepSeek-V3-0324 |
模型简介 | Not provided | Not provided | Not provided |
模型类型 | 聊天大模型 | 推理大模型 | 聊天大模型 |
模型代号 | deepseek-v-3_1 | DeepSeek-R1-0528 | DeepSeek-V3-0324 |
Release | 2025-08-20 | 2025-05-28 | 2025-03-24 |
MoE | Yes | Yes | Yes |
规格与性能 | |||
Context length | 128K | 64K | 128K |
Parameters | 6710 | 6710 | 6710 |
激活参数量 | 370 | 370 | 370 |
模型规模 | 100b | 100b | 100b |
模型大小 | 1340GB | 685GB | 1442GB |
推理速度 | |||
推理等级 | |||
最大输出 | 8192 | 64000 | Not provided |
Supported modes | 常规模式(Non-Thinking Mode)思考模式(Thinking Mode) | 思考模式(Thinking Mode) | 常规模式(Non-Thinking Mode) |
开源与许可 | |||
Code Open Source | Closed Source | Closed Source | Closed Source |
Weights Open Source | Closed Source | Closed Source | Closed Source |
Commercial use | 免费商用授权 | 免费商用授权 | 免费商用授权 |
Modality support | |||
Text Input/Output | / | / | / |
Image Input/Output | / | / | / |
Audio Input/Output | / | / | / |
Video Input/Output | / | / | / |
Embedding Input/Output | / | / | / |
API 接口详情 | |||
Text 价格 | Input: 0.56 美元/100 万tokensOutput: 1.68 美元/100 万tokens | Input: 0.55 美元/ 100 万tokensOutput: 2.19 美元/ 100 万tokens | Input: 0.27 美元/100万 tokensOutput: 1.1 美元/100万 tokens |
Image API pricing | Not provided | Not provided | Not provided |
Audio API pricing | Not provided | Not provided | Not provided |
Video API pricing | Not provided | Not provided | Not provided |
Embedding API pricing | Not provided | Not provided | Not provided |
Resources | |||
GitHub | Not provided | Repo | Repo |
Hugging Face | Model Page | Model Page | Model Page |
Official Page | Not provided | Not provided | Not provided |
Guides | Not provided | Not provided | Not provided |
Papers | DeepSeek-V3.1 Release | DeepSeek-R1-0528 Release | |
DataLearnerAI | DeepSeek V4没有等到,但是DeepSeekAI把DeepSeek V3升级到DeepSeek V3.1了,小幅更新,但核心架构和参数不变 | Not provided | DeepSeekV3-0324发布:DeepSeek V3基础上大幅升级推理能力和前端网页的美观度,多项评测结果超过GPT-4.5 |
API pricing
Side-by-side input/output token pricing