Minimax

MiniMax M3

Chat
Not available
Get started

Sparse MoE with MiniMax Sparse Attention (MSA). 15.6x faster decoding at 1M context. Multimodal (text, image, video).

Context
1Mtokens
Max output
131Ktokens
Latency
~14ms
Released
2026-05-31

Pricing

$ / 1M tokens
Input
$0.30/M
Output
$1.20/M
Cache Read
$0.06/M
Cache Write
$0.30/M

Capabilities

3 listed
  • Chat
  • Tool Use
  • Coding

Code examples

POST /v1/chat/completions
python
from openai import OpenAI

client = OpenAI(
    api_key="your_api_key",
    base_url="https://api.modelstack.cc/v1",
)

response = client.chat.completions.create(
    model="minimax-m3",
    messages=[{"role": "user", "content": "Hello!"}],
)
print(response.choices[0].message.content)

Related models

Frequently asked questions

How is MiniMax M3 billed on ModelStack?expand_more
ModelStack bills based on actual request usage. Input, output, cache, and any other applicable rates follow the live pricing table on this page. Final request costs are available in the console.
What are the context window and maximum output for MiniMax M3?expand_more
MiniMax M3 has a 1M context window and supports up to 131K of output.
How can I call MiniMax M3 through ModelStack?expand_more
Use the model ID `minimax-m3` with any OpenAI-compatible SDK pointed at the ModelStack API — see the code examples above for Python, Node.js, and cURL.
Which capabilities does MiniMax M3 support?expand_more
MiniMax M3 supports: Chat, Tool Use, Coding.