See key specs and per-benchmark scores for each model/mode. Scroll horizontally for all columns. 当前对比 2 个模型的评测数据与核心参数。
DeepSeek V4 Pro
DeepSeek-V3.1
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-V4-Pro · 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-V4-ProDeepSeek-AI | DE DeepSeek-V3.1DeepSeek-AI |
|---|---|---|
Model snapshots | ||
Organization | DeepSeek-AI | DeepSeek-AI |
模型全名 | DeepSeek V4 Pro | DeepSeek-V3.1 |
模型简介 | Not provided | Not provided |
模型类型 | 推理大模型 | 聊天大模型 |
模型代号 | deepseek-v4-pro | deepseek-v-3_1 |
Release | 2026-04-24 | 2025-08-20 |
MoE | Yes | Yes |
规格与性能 | ||
Context length | 1M | 128K |
Parameters | 16000 | 6710 |
激活参数量 | 490 | 370 |
模型规模 | 100b | 100b |
模型大小 | Not provided | 1340GB |
推理速度 | ||
推理等级 | ||
最大输出 | 384000 | 8192 |
Supported modes | No mode data | 常规模式（Non-Thinking Mode）思考模式（Thinking Mode） |
开源与许可 | ||
Code Open Source | Closed Source | Closed Source |
Weights Open Source | Closed Source | Closed Source |
Commercial use | 免费商用授权 | 免费商用授权 |
Modality support | ||
Text Input/Output | / | / |
Image Input/Output | Not provided | / |
Audio Input/Output | Not provided | / |
Video Input/Output | Not provided | / |
Embedding Input/Output | Not provided | / |
API 接口详情 | ||
Text 价格 | Input: $1.74 / 1M tokensOutput: $3.48 / 1M tokensCache: $1.74 / 1M tokens | Input: 0.56 美元/100 万tokensOutput: 1.68 美元/100 万tokens |
Image API pricing | Not provided | Not provided |
Audio API pricing | Not provided | Not provided |
Video API pricing | Not provided | Not provided |
Embedding API pricing | Not provided | Not provided |
Resources | ||
GitHub | Not provided | Not provided |
Hugging Face | Model Page | Model Page |
Official Page | Not provided | Not provided |
Guides | Not provided | Not provided |
Papers | DeepSeek-V4 Technical Report | DeepSeek-V3.1 Release |
DataLearnerAI | Not provided | 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 |
API pricing
Side-by-side input/output token pricing
