metadata
license: mit
pipeline_tag: text-generation
tags:
- cortex.cpp
Overview
DeepSeek developed and released the DeepSeek R1 Distill Llama 70B model, a distilled version of the Llama 70B language model. This model represents the pinnacle of the DeepSeek R1 Distill series, designed for exceptional performance in text generation, dialogue tasks, and advanced reasoning, offering unparalleled capabilities for large-scale AI applications.
The model is ideal for enterprise-grade applications, research, conversational AI, and large-scale knowledge systems, providing top-tier accuracy, safety, and efficiency.
Variants
| No | Variant | Cortex CLI command |
|---|---|---|
| 1 | Deepseek-r1-distill-llama-70b-70b | cortex run deepseek-r1-distill-llama-70b:70b |
Use it with Jan (UI)
- Install Jan using Quickstart
- Use in Jan model Hub:
cortexso/deepseek-r1-distill-llama-70b
Use it with Cortex (CLI)
- Install Cortex using Quickstart
- Run the model with command:
cortex run deepseek-r1-distill-llama-70b
Credits
- Author: DeepSeek
- Converter: Homebrew
- Original License: License
- Papers: DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning