cortexso/deepseek-r1-distill-qwen-14b

Overview

DeepSeek developed and released the DeepSeek R1 Distill Qwen 14B model, a distilled version of the Qwen 14B language model. This variant represents the largest and most powerful model in the DeepSeek R1 Distill series, fine-tuned for high-performance text generation, dialogue optimization, and advanced reasoning tasks.

The model is designed for applications that require extensive understanding, such as conversational AI, research, large-scale knowledge systems, and customer service, providing superior performance in accuracy, efficiency, and safety.

Variants

Use it with Jan (UI)

1. 1.Install Jan using Quickstart
2. 2.Use in Jan model Hub:

    cortexso/deepseek-r1-distill-qwen-14b

Use it with Cortex (CLI)

1. 1.Install Cortex using Quickstart
2. 2.Run the model with command:

    cortex run deepseek-r1-distill-qwen-14b

Credits

• —Author: DeepSeek
• —Converter: Homebrew
• —Original License: License
• —Papers: DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning