← Back to AI Models

DeepSeek V2

Efficient MoE model with strong general capabilities

Overview

Efficient MoE model with strong general capabilities The model has 236B (21B active) parameters. It is an open-weight model released under the DeepSeek License licence.

Key Capabilities

  • Code Generation — suited for tasks requiring code generation proficiency
  • Reasoning — suited for tasks requiring reasoning proficiency
  • Chat — suited for tasks requiring chat proficiency
  • Analysis — suited for tasks requiring analysis proficiency

Recommended Use Cases

  • Cost Effective Inference
  • General Tasks
  • Self Hosting

Technical Specifications

  • Context Window: 128K tokens
  • Parameters: 236B (21B active)
  • Open Weight: Yes
  • Licence: DeepSeek License
  • Release Date: May 06, 2024

Provider

DeepSeek V2 is developed by DeepSeek. For the latest information, visit the official documentation.