---
id: "minimax/minimax-m2.1"
also: ["minimax/minimax-m2.1", "MiniMaxAI/MiniMax-M2.1", "hf/MiniMaxAI/MiniMax-M2.1", "vercel/minimax/minimax-m2.1"]
card_usd_per_m: "0.3 in, 1.2 out, at minimax/fp8"
cheapest_list_usd_per_m: "0.3 in, 1.2 out, at novita on hf"
cheapest_effective_usd_per_m: "0.0952 in, 1.1994 out, at minimax/fp8 on openrouter"
context: 204800
max_output: 131072
input: ["text"]
output: ["text"]
released: "2025-12-23"
params: ["frequency_penalty", "include_reasoning", "max_tokens", "presence_penalty", "reasoning", "repetition_penalty", "response_format", "seed", "stop", "structured_outputs", "temperature", "tool_choice", "tools", "top_k", "top_p"]
endpoints: 8
gateways: "openrouter (3), vercel (3), hf (2)"
no_training: 4
trains: 0
retains_prompts: 0
intelligence: 20.9000
---
# MiniMax: MiniMax M2.1

MiniMax-M2.1 is a lightweight, state-of-the-art large language model optimized for coding, agentic workflows, and modern application development. With only 10 billion activated parameters, it delivers a major jump in real-world...