Is Llama 3.1 405B Instruct or MiniMax-M2.1 cheaper?

Llama 3.1 405B Instruct is cheaper on output tokens ($0.00 vs $1.34 per 1M).

Which has the larger context window, Llama 3.1 405B Instruct or MiniMax-M2.1?

MiniMax-M2.1 has the larger context window (196K tokens).

Llama 3.1 405B Instruct vs MiniMax-M2.1

Llama 3.1 405B Instruct is cheaper on output tokens, while MiniMax-M2.1 offers a larger context window. Choose Llama 3.1 405B Instruct or MiniMax-M2.1 based on the trade-off between cost, context, and the benchmarks that matter for your use case.

Spec	Llama 3.1 405B Instruct	MiniMax-M2.1
Provider	Cortecs	Cortecs
Input / 1M tokens	$0.00	$0.34
Output / 1M tokens	$0.00	$1.34
Context window	128K	196K
Parameters	—	229B
Open weights	Yes	Yes
Released	Jul 2024	Dec 2025

Llama 3.1 405B Instruct details →MiniMax-M2.1 details →

FAQ

Pricing is indicative — confirm with the provider before production use. Highlighted values indicate the better figure for that row.