RELEASE · MODELS · #2118
Reflection releases Beam — 501B MoE open model with 1M-token context (Apache‑2.0)
Reflection published Beam, a 501B-parameter mixture-of-experts (MoE) language model on Hugging Face under the Apache‑2.0 license. Beam runs ~23B parameters per token, supports text input with a 1M-token context window, targets coding and agentic use cases, ships with a Rust performance harness and day‑0 support for transformers, vLLM and SGLang, and includes recipes for inference, SFT and RL.
KEY POINTS
- Reflection published Beam, a 501B-parameter mixture-of-experts (MoE) language model on Hugging Face under the Apache‑2.0 license.
- Beam runs ~23B parameters per token, supports text input with a 1M-token context window, targets coding and agentic use cases, ships with a Rust performance harness and day‑0 support for transformers, vLLM and SGLang, and includes recipes for inference, SFT and RL.
- An open, frontier-scale MoE with 1M context and day‑0 ecosystem support lowers barriers for teams to run, customize, and research large long-context agentic models on their own infrastructure.
WHY IT MATTERS
An open, frontier-scale MoE with 1M context and day‑0 ecosystem support lowers barriers for teams to run, customize, and research large long-context agentic models on their own infrastructure.