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
MiniMax's fast multimodal agent and coding model with tunable thinking depth, the generally available successor of the M3.1 Flash Preview. it costs half of MiniMax M3's sub-512K per-token rate.
pick it if
you want MiniMax at half of M3's sub-512K per-token rate
you send prompts past 512K tokens; the rate stays flat
your agent re-reads long context; cached input bills at a tenth
skip it if
you need JSON-schema constrained output
you need downloadable weights today; MiniMax says they come later
faq
which is better, Beam or MiniMax M3.1 Flash?
on DeepSWE v1.1, as each maker reports it, Beam scores 43.3% and MiniMax M3.1 Flash 73.7%. these are each maker's own numbers, not a controlled head-to-head; the spec table lists every benchmark both publish.
which is cheaper, Beam or MiniMax M3.1 Flash?
Beam lists lower input and output prices. on a coding-agent session (2M tokens in, 60,000 out) Beam costs $0.27 and MiniMax M3.1 Flash $0.34.
which one accepts images?
MiniMax M3.1 Flash supports vision; Beam does not.
can i switch between Beam and MiniMax M3.1 Flash mid-task?
yes. in an octomind session, /model octohub:beam and /model octohub:minimax-m3.1-flash 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.