MiniMax-M2.7 and M2.7-highspeed API

MiniMax's cost-efficient chat and coding models from $0.3134 per million input tokens, reachable through OpenAI-compatible and Anthropic-compatible endpoints.

MiniMax-M2.7 and M2.7-highspeed

MiniMax-M2.7 is the latest release in the MiniMax M2 family, aimed at chat, coding and agent workloads with a strong cost-to-quality ratio. M2.7-highspeed is the same model served with higher throughput for faster first-token and end-to-end latency, ideal for interactive applications.

MiniMax-M2.7 API pricing (USD per 1M tokens)

Model	Input	Output	Cache read	Cache write
MiniMax-M2.7	$0.3134	$1.2537	$0.0627	$0.3918
MiniMax-M2.7-highspeed	$0.6269	$2.5075	$0.0627	$0.3918

Quick start

curl

curl https://yiduochan.com/v1/chat/completions \
  -H "Authorization: Bearer $YIDUOCHAN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "MiniMax-M2.7",
    "messages": [{"role": "user", "content": "Introduce the MiniMax models in one sentence"}]
  }'

Python (OpenAI SDK)

from openai import OpenAI

client = OpenAI(base_url="https://yiduochan.com/v1", api_key="YIDUOCHAN_API_KEY")
resp = client.chat.completions.create(
    model="MiniMax-M2.7-highspeed",
    messages=[{"role": "user", "content": "Introduce the MiniMax models in one sentence"}],
)
print(resp.choices[0].message.content)

Using it in popular tools

Cursor, Continue, Cline: choose an OpenAI-compatible provider, set the base URL to https://yiduochan.com/v1 and the model to MiniMax-M2.7.
Cherry Studio, NextChat, LobeChat: add an OpenAI-type provider with your key and enter the model name manually.
LangChain / LlamaIndex: use ChatOpenAI with base_url and api_key.

Use it in Claude Code and other Anthropic-compatible tools

The same key also works against our Anthropic Messages compatible endpoint /v1/messages, so MiniMax models plug straight into Claude Code or the Anthropic mode of Cline. For Claude Code, export three variables:

export ANTHROPIC_BASE_URL=https://yiduochan.com
export ANTHROPIC_AUTH_TOKEN=YIDUOCHAN_API_KEY
export ANTHROPIC_MODEL=MiniMax-M2.7

About reasoning output

The M2.7 family emits its reasoning. Over the OpenAI-compatible endpoint the thinking is wrapped in <think> tags inside the message content; reasoning_tokens is reported separately in usage and billed at the output rate.

FAQ

What is the difference between MiniMax-M2.7 and M2.7-highspeed?

Same model, different serving: highspeed answers faster at twice the price, which suits latency-sensitive chat and code completion.

What does the MiniMax-M2.7 API cost?

The standard model is $0.3134 input and $1.2537 output per million tokens, with $0.0627 for cache reads and $0.3918 for cache writes; highspeed is $0.6269 input and $2.5075 output.

Can I use MiniMax-M2.7 in Cursor or Cline?

Yes. Any tool that accepts a custom OpenAI-compatible base URL works: set it to https://yiduochan.com/v1 and use a key from this site.

How do I run Claude Code on MiniMax-M2.7?

Export ANTHROPIC_BASE_URL=https://yiduochan.com, ANTHROPIC_AUTH_TOKEN with your key and ANTHROPIC_MODEL=MiniMax-M2.7, then start Claude Code; requests are forwarded through the Anthropic Messages compatible endpoint.