minimax/minimax-m3

MiniMax M3

MiniMax-M3 is a multimodal foundation model from MiniMax. It supports text, image, and video inputs with text output, a 1M-token context window, and is suited for long-horizon agentic work, coding,...

5/31/2026

1,048,576 tokens

#23 Code (Overall)

Input: ¥53.6145/M

Output: ¥214.4579/M

Try in Chat Get API Key

Specifications

Modalities

Input

text

image

video

Output

text

Supported Parameters

frequency_penalty

include_reasoning

logit_bias

logprobs

max_tokens

min_p

presence_penalty

reasoning

repetition_penalty

response_format

seed

stop

structured_outputs

temperature

tool_choice

tools

top_k

top_logprobs

top_p

Max Output Tokens

512,000

Code Examples

curl https://api.fastmetal.ai/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "model": "minimax-m3",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

Leaderboard

Text

🏆OverallELO: 1,444

#64

🇯🇵JapaneseELO: 1,417

#46

🇨🇳ChineseELO: 1,475

#79

🇰🇷KoreanELO: 1,370

#84

🇬🇧EnglishELO: 1,457

#59

frenchELO: 1,456

#72

germanELO: 1,429

#72

spanishELO: 1,422

#79

russianELO: 1,440

#68

💻CodingELO: 1,498

#54

🧮MathELO: 1,443

#60

✍️Creative WritingELO: 1,407

#71

📝Instruction FollowingELO: 1,437

#57

🌶️Hard PromptsELO: 1,464

#62

💬Multi-TurnELO: 1,452

#62

Vision

🏆OverallELO: 1,240

#38

Code

🏆OverallELO: 1,491

#23