Skip to main content

reflection.ai

2 related articles

Subscribe to updates

reflection.ai Oct 16, 2026
Reflection opens 501B-parameter Beam weights under Apache 2.0, with 40-plus partners turning one model into a self-hostable stack
Reflection released Beam, its first open-weight model, with 501 billion total parameters and 23 billion active in a sparse Mixture-of-Experts design for coding, reasoning and long-horizon agentic workloads, publishing weights on Hugging Face under Apache 2.0 with a model card and a safety technical report. Where an open-weight release usually delivers one model layer, this one also ships BF16, FP8 and NVFP4 precision formats, a speculative decoding drafter, and integrations from more than 40 partners across inference, fine-tuning, deployment and agent harnesses, so developers can download, self-host and customize it for their own domains.
Reflection released Beam, its first open-weight model, with 501 billion total parameters and 23 billion active in a sparse Mixture-of-Experts design for coding, reasoning and long-horizon agentic workloads, publishing weights on Hugging Face under Apache 2.0 with a model card and a safety technical report. Where an open-weight release usually delivers one model layer, this one also ships BF16, FP8 and NVFP4 precision formats, a speculative decoding drafter, and integrations from more than 40 partners across inference, fine-tuning, deployment and agent harnesses, so developers can download, self-host and customize it for their own domains.
reflection.ai
Beam Mixture-of-Experts MIRROR Reinforcement Learning
Baseten Oct 16, 2026
Reflection opens Beam, its first open-weight model: 501B parameters, 23B active, fewer tokens per coding task
Reflection released Beam, its first open-weight model, with day-0 support on the Baseten platform. Beam is an autoregressive transformer with a mixture-of-experts architecture: 501B total parameters, 23B active per token, and a context window up to 1 million tokens, aimed at coding and agentic workloads. Training leaned on reinforcement learning, using longer multi-step rollouts so the model can practice tool use and adjust from feedback, plus a length penalty against unnecessary tokens, with the goal of cutting the cost of these workloads.
Reflection released Beam, its first open-weight model, with day-0 support on the Baseten platform. Beam is an autoregressive transformer with a mixture-of-experts architecture: 501B total parameters, 23B active per token, and a context window up to 1 million tokens, aimed at coding and agentic workloads. Training leaned on reinforcement learning, using longer multi-step rollouts so the model can practice tool use and adjust from feedback, plus a length penalty against unnecessary tokens, with the goal of cutting the cost of these workloads.
reflection.ai
Beam Reinforcement Learning Mixture-of-Experts

You’re all caught up