Introducing Reflection Beam, an open-weight model for agentic coding on AWS
Introducing Reflection Beam: an open-weight model for agentic coding on AWS
- Who
- AWS
- What
- explains how to deploy Reflection Beam on SageMaker JumpStart and HyperPod
- Quote
In this post, we explain Reflection’s vision and show how you can deploy Beam on AWS.
What happened
Reflection AI introduced Beam, an open-weight model for agentic coding and multi-step tasks, with 501 billion total parameters and 23 billion active parameters. Reflection reports an effective context length of 1 million tokens and releases Beam 1.0 under Apache 2.0 in BF16, FP8, and NVFP4 variants. The guide explains deployment through Amazon SageMaker JumpStart or HyperPod Inference with JumpStart, SGLang, or vLLM, including capacity planning, invocation, performance measurement, and cleanup.
Why it matters
The deployment comparison clarifies when managed endpoints suffice and when Kubernetes infrastructure or custom runtimes offer useful control over serving configuration, distributed execution, and scaling.
The story so far
Reflection AI introduced Beam 1.0, an open-weight model for agentic coding and multi-step tasks, and AWS published instructions for deploying it through Amazon SageMaker JumpStart or HyperPod Inference. At launch, Beam is available in JumpStart in US East (N. Virginia) and US East (Ohio). Beam has 501 billion total parameters and 23 billion active parameters. Reflection reports an effective context length of 1 million tokens and released the model under Apache 2.0 in BF16, FP8, and NVFP4 variants. AWS’s guide covers capacity planning, invocation, performance measurement, and cleanup, with deployment options using JumpStart, SGLang, or vLLM. For FP8 deployment, AWS specifies ml.p5.48xlarge instances with NVIDIA H100 GPUs or ml.p5e.48xlarge and ml.p5en.48xlarge instances with H200 GPUs; for NVFP4, it specifies ml.p6-b200.48xlarge instances with B200 GPUs.
Keep reading
Latest daily · Issue 24NVIDIA and AI4Bharat Introduce Nemotron-pretraining-Indic-v1 and an Open Data Generation PipelineTrending now
- 1Meta Releases Muse Spark Open With Mac and Cloud Support, Topping US Open-Weight Index
- 2Google Launches Gemini 4 Argon Globally, Expands Paid Access as Benchmark Results Vary
- 3OpenAI Updates GPT-6 Astra Pro in ChatGPT With Parallel Agents and Faster Responses
- 4Meta’s New Business Terms Raise Dispute Over Website Crawling and AI Training Rights