Reflection

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...

beam-1.0-501b-a23b

Pricing

~1%

Input: $0.297 / 1,000,000 tokens
Output: $0.99 / 1,000,000 tokens
Cache read: $0.0396 / 1,000,000 tokens
Cache write: $0.297 / 1,000,000 tokens

How pricing works

Discount history

1%Nov 8, 03:58 AM UTC

Nov 7, 11:59 PMNov 8, 03:58 AM UTC

Model details

Input: text
Output: text
Context window: 1,048,576 tokens
Maximum output: 524,288 tokens
Availability: Uncapped capacity
Updated: 11/8/2026, 4:00:11 AM UTC

Capacity is reserved when your request is accepted.

Capabilities

ChatStreamingTool callingReasoningJSON mode

Supported parameters

frequency_penaltyinclude_reasoninglogit_biaslogprobsmax_tokensmin_ppresence_penaltyreasoningreasoning_effortrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_logprobstop_p