Open Source LLMs
DeepSeek-V3
🇨🇳 中文 / 🇬🇧 English — 双语介绍
🇨🇳 中文
深度求索开源的 671B 参数 MoE 大模型，每个 token 只激活约 37B 参数，把推理成本压得很低。权重、推理代码与技术报告全部公开，训练成本量级与工程细节透明，是开源阵营里被复现和二次训练最多的基座之一。 🔗 GitHub Stars: 104k+
🇬🇧 English
DeepSeek's 671B-parameter MoE model that activates roughly 37B parameters per token, cutting inference cost sharply. Weights, inference code and a detailed technical report are all public, which made it one of the most reproduced open bases. 🔗 GitHub Stars: 104k+
📋 Product Info
Category: Open Source LLMs
Listed on: 2026-10-12
Official site: https://github.com/deepseek-ai/DeepSeek-V3
Open source: GitHub repository ↗
Tags: llm,open-source,moE,deepseek,reasoning
🎯 Why This Tool
DeepSeek-V3 belongs to the Open Source LLMs category on hedirbase. Tagged with: llm,open-source,moE,deepseek,reasoning.
DeepSeek's 671B-parameter MoE model that activates roughly 37B parameters per token, cutting inference cost sharply. Weights, inference code and a detailed technical report are all public, which made
✓ Manually reviewed
📁 Open source
🌐 Curated for independents