Today, we are excited to announce that DeepSeek R1 distilled Llama and Qwen designs are available through Amazon Bedrock Marketplace and Amazon SageMaker JumpStart. With this launch, you can now release DeepSeek AI’s first-generation frontier model, DeepSeek-R1, in addition to the distilled versions varying from 1.5 to 70 billion specifications to build, experiment, and responsibly scale your generative AI ideas on AWS.

In this post, we show how to get going with DeepSeek-R1 on Amazon Bedrock Marketplace and SageMaker JumpStart. You can follow comparable steps to release the distilled variations of the models also.

Overview of DeepSeek-R1

DeepSeek-R1 is a big language model (LLM) established by DeepSeek AI that uses reinforcement learning to boost thinking abilities through a multi-stage training procedure from a DeepSeek-V3-Base foundation. A crucial distinguishing function is its support knowing (RL) step, which was used to fine-tune the design’s reactions beyond the standard pre-training and fine-tuning process. By including RL, DeepSeek-R1 can adjust better to user feedback and goals, eventually enhancing both relevance and clearness. In addition, DeepSeek-R1 utilizes a chain-of-thought (CoT) method, indicating it’s geared up to break down complex inquiries and factor through them in a detailed way. This guided reasoning process permits the model to produce more accurate, transparent, and detailed answers. This design combines RL-based fine-tuning with CoT capabilities, aiming to create structured actions while concentrating on interpretability and user interaction. With its wide-ranging capabilities DeepSeek-R1 has recorded the market’s attention as a versatile text-generation design that can be integrated into different workflows such as agents, sensible reasoning and data analysis tasks.

DeepSeek-R1 uses a Mixture of Experts (MoE) architecture and yewiki.org is 671 billion parameters in size. The MoE architecture permits activation of 37 billion parameters, allowing efficient reasoning by routing questions to the most pertinent expert “clusters.” This technique allows the model to concentrate on various problem domains while maintaining general performance. DeepSeek-R1 requires at least 800 GB of HBM memory in FP8 format for reasoning. In this post, we will use an ml.p5e.48 xlarge circumstances to release the model. ml.p5e.48 xlarge comes with 8 Nvidia H200 GPUs offering 1128 GB of GPU memory.

DeepSeek-R1 distilled designs bring the thinking abilities of the main R1 model to more effective architectures based upon popular open models like Qwen (1.5 B, 7B, 14B, and 32B) and Llama (8B and 70B). Distillation describes a procedure of training smaller, more effective models to mimic the habits and reasoning patterns of the bigger DeepSeek-R1 model, using it as an instructor model.

You can release DeepSeek-R1 design either through SageMaker JumpStart or Bedrock Marketplace. Because DeepSeek-R1 is an emerging design, we recommend deploying this design with guardrails in place. In this blog, [forum.batman.gainedge.org](https://forum.batman.gainedge.org/index.php?action=profile