Verified65% confidenceFactTime unknown

DeepSeek-V3在多项评测中性能超越了通义千问2.5和Llama 3.1等开源模型

3

Sources

65%

Confidence

Medium-term (~90 days)

Relevance

6/2/2026

First Seen

Valid until: 8/31/2026

Sources

DeepSeek-V3+VSCode搭建低成本AI代码编辑器教程

bilibiliLLM大模型应用开发

Related Entities

DeepSeek-V3model 通义千问2.5model Llama 3.1model

Related Claims

UnverifiedDeepSeek开放平台提供两个主要级别的模型：DeepSeek V3（响应速度快、Token成本低）和DeepSeek R1系列（推理增强版）80% similar VerifiedDeepSeek发布了V3.2版本模型，在多项基准测试中追平Gemini 3.0 Pro79% similar UnverifiedDeepSeek V4在智能体（Agentic Capabilities）评测中排名前三79% similar

Cite This Claim

Stable URI

https://kongchang.com/claim/37186

API

curl https://kongchang.com/api/v1/knowledge/claims/37186

MCP

get_claim(id=37186)