MiniMax M3 在 NextModel 中是什麼?
香港團隊可用的 NextModel 目錄中的 MiniMax 模型,常用於 high-volume chat、agentic tool use、分類 工作負載。當前展示價格為輸入 ¥2.81 / 1M tokens、輸出 ¥11.23 / 1M tokens 每 100 萬 token,上下文視窗為 128k token。
適用場景
- high-volume chat
- agentic tool use
- 分類
OpenAI 相容呼叫範例
保持 OpenAI SDK 呼叫方式不變,把 base_url 改為 NextModel,並使用模型目錄 ID minimax-m3。
from openai import OpenAI
client = OpenAI(
api_key="YOUR_API_KEY",
base_url="https://api.nextmodel.app/v1"
)
resp = client.chat.completions.create(
model="minimax-m3",
messages=[{"role": "user", "content": "Hello from NextModel"}]
)
print(resp.choices[0].message.content)相似替代項
MiniMax M2.7 is available only through the Volcengine Agent Plan (no public ARK price list); its listed price tracks the OpenRouter reference rate for the same model. Unlike M3, this model has no cache-hit discount.
DeepSeek V4 Flash 是長上下文、低成本候選,適合需要中文能力與批量友善定價的團隊。
Doubao Seed 2.0 Lite is the low-cost long-context member of the Seed 2.0 family, onboarded through the Volcengine Agent Plan. Audio input pricing is not modeled in this catalog.
常見問題
MiniMax M3 API 常見問題
What is MiniMax M3 best for?
Cost-sensitive high-volume chat, classification, and agentic tool use.
How is MiniMax M3 priced through NextModel?
¥2.81 per 1M input tokens and ¥11.23 per 1M output tokens, tracking MiniMax's OpenRouter reference rate.