Deepseek Coder V2 Instruct
DeepSeek develops the Deepseek Coder V 2 Instruct, a chat model exceling at code-related tasks such as code completion, generation, and review, thanks to its substantial pre-training with an additional 6 trillion tokens. Notably, it utilizes a Mixture-of-Experts (MoE) architecture, allowing for efficient use of parameters.
Input
Output
Context
131K
Max Output
66K
Parameters
235.7B
Technical Specifications
Model TypeChat
Context Window131,072 tokens
Max Output Tokens65,536 tokens
Parameters235.7B
Release DateJun 14, 2024
Training CutoffNot available
Licenseother
Open SourceYes
Input Modalities
Text
Output Modalities
Text
Capabilities
Resources & Links
Estimated GPU Requirements for Deepseek Coder V2 Instruct
Estimates based on INT8 quantization. Actual requirements vary by framework and configuration.
Related Tools
Data sourced from official provider APIs and documentation
Last updated: Jun 23, 2026
Start building with the right model.
Automatically route workloads to the right model for every task, every time.