the first open-weights model from Reflection, a 501B-total, 23B-active mixture-of-experts built for agentic tool use, coding and general reasoning. it reads text only and its weights ship under Apache 2.0.
pick it if
you want a long-window text model priced with the flash tier
you want Apache 2.0 weights, with BF16, FP8 and NVFP4 builds to self-host
you run tool-calling agents on text and repeat long prompts, so cached input matters
skip it if
you need image, audio or video input; Beam reads text only
long autonomous coding is the job; its reported DeepSWE v1.1 score sits below GLM-5.3 Flash and DeepSeek V4.1 Flash
GPT-6 Luna is the lowest-priced tier of OpenAI's GPT-6 family, below Sol and Astra; OpenAI calls it its most efficient model for focused, high-volume tasks. it accepts text and image input and replaced GPT 5.6 Luna at half the input price.
pick it if
you run high-volume tasks at the GPT-6 family's lowest price
you resend long prompts: cached input is $0.01 per 1M tokens
you need image input and structured outputs on a budget
skip it if
your hardest tasks need a stronger tier such as GPT-6.1 Sol or Astra
most requests exceed 272K input tokens, where rates rise
faq
which is cheaper, Beam or GPT-6 Luna?
GPT-6 Luna lists the lower input price, while Beam lists the lower output price. on a coding-agent session (2M tokens in, 60,000 out) Beam costs $0.27 and GPT-6 Luna $0.23.
which one accepts images?
GPT-6 Luna supports vision; Beam does not.
can i switch between Beam and GPT-6 Luna mid-task?
yes. in an octomind session, /model octohub:beam and /model octohub:gpt-6-luna switch models without losing the conversation, so you can draft on one and review on the other.
try either model
switch between models inside the same octomind session.