Skip to content
CheapestInference

MiniMax M3 API — unlimited & flat-rate access

MiniMax M3 is MiniMax’s frontier multimodal coding and agentic model. CheapestInference serves it through an OpenAI- and Anthropic-compatible API on unlimited, flat-rate time-block subscriptions — so your cost does not scale with tokens.

Quick facts

Section titled “Quick facts”
ModelMiniMax M3
Model IDminimax-m3
Context window1M tokens
Max request size8 MB per request
Per-token price elsewhere$0.30 in / $1.20 out per 1M — reference only
Endpoints/v1/chat/completions (OpenAI), /anthropic/v1/messages (Anthropic)
PricingFrom $71/mo — reserve an 8-hour daily time block, up to full 24/7

Call MiniMax M3

Section titled “Call MiniMax M3”
from openai import OpenAI


client = OpenAI(
    base_url="https://api.cheapestinference.com/v1",
    api_key="sk-..."  # your subscriber key
)


response = client.chat.completions.create(
    model="minimax-m3",
    messages=[{"role": "user", "content": "Summarize this document..."}],
)
Terminal window
curl https://api.cheapestinference.com/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "minimax-m3", "messages": [{"role": "user", "content": "Hello"}]}'

Why flat-rate MiniMax M3

Section titled “Why flat-rate MiniMax M3”

MiniMax M3 pairs frontier coding and agentic ability with multimodal input, making it well suited to large codebases, long documents, and long-running agent loops. On CheapestInference it is billed at a flat monthly rate, not per token, so heavy workloads have a predictable cost. It is part of the Frontier Pool alongside Kimi K2.7 and GLM 5.3, so one subscription covers all three. It works in any OpenAI-compatible client.

Common questions

Section titled “Common questions”

Is there a MiniMax M3 API? Yes. Use model id minimax-m3 against https://api.cheapestinference.com/v1. The API is OpenAI- and Anthropic-SDK compatible.

How much does MiniMax M3 cost? From $71/month ($60.35/mo billed annually). You reserve one or more 8-hour daily time blocks (up to full 24/7) and use MiniMax M3 with no token caps — billed at a flat monthly fee, not per token.

How large can a request be? Each request to the Frontier Pool accepts up to 8 MB of payload — messages, history and attachments combined. Larger requests are rejected with a context_length_exceeded error; agent clients such as Claude Code compact the conversation automatically and retry. See Request size limits.

Is MiniMax M3 good for coding? Yes — it is a frontier coding and agentic model, and is served alongside GLM 5.3 in the Frontier Pool.

Section titled “Related”