---
id: "reflection/beam-1.0-501b-a23b"
also: ["reflection/beam-1.0-501b-a23b-20261016", "reflection/Beam-1.0-501B-A23B", "hf/reflection/Beam-1.0-501B-A23B"]
cheapest_list_usd_per_m: "0.25 in, 1.0 out, at nebius/fp4 on openrouter"
cheapest_effective_usd_per_m: "0.0612 in, 1.1994 out, at together on openrouter"
context: 1048576
max_output: 943718
input: ["text"]
output: ["text"]
released: "2026-10-16"
reasoning: ["max", "xhigh", "high", "medium", "low"]
api: ["/api/v1/chat/completions", "/api/v1/completions", "/api/v1/messages", "/api/v1/responses"]
params: ["frequency_penalty", "include_reasoning", "logit_bias", "logprobs", "max_tokens", "min_p", "presence_penalty", "reasoning", "reasoning_effort", "repetition_penalty", "response_format", "seed", "stop", "structured_outputs", "temperature", "tool_choice", "tools", "top_k", "top_logprobs", "top_p"]
endpoints: 6
gateways: "openrouter (5), hf (1)"
no_training: 5
trains: 0
retains_prompts: 0
intelligence: 17.7000
---
# Reflection: Beam 1.0 501B A23B

Beam 1.0 is Reflection's open-weight reasoning model for coding and long-horizon agentic workflows. It uses a sparse mixture-of-experts architecture with 501B total and 23B active parameters per token, and a...