How much does the MiniMax M3 API cost?

On FastMetal, MiniMax M3 is billed per token in yen: ¥53.61 per 1M input tokens and ¥214.44 per 1M output tokens, before tax. Usage is drawn from a prepaid balance; there is no subscription or monthly fee.

Can I call MiniMax M3 with the OpenAI SDK?

Yes. Point base_url at https://api.fastmetal.ai/v1 and pass "minimax-m3" as the model; existing OpenAI-style code works unchanged, including streaming, tool calls and structured output.

What is the context window of MiniMax M3?

1,048,576 tokens, shared between the prompt and the response.

How does MiniMax M3 rank?

#74 on the public arena's Japanese board (ELO 1,398). Ranks move as the leaderboard is updated.

What do I need to try MiniMax M3?

Create an account and add credit; the model is then available both in the browser chat and over the API. There is no contract or minimum spend.

Back to Models

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

#47 Code (Overall)

Input: ¥53.61/M

Output: ¥214.44/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!"}]
  }'

How MiniMax M3 actually answers

Real responses to our standard prompts, recorded on FastMetal.

Count the number of 'r's in 'strawberry'

Count the number of 'r's in 'strawberry'. Explain your reasoning step by step.

# Counting 'r's in "strawberry" Let me go through the word letter by letter: **s** - t - **r** - a - w - b - e - **r** - **r** - y | Position | Letter | Is it 'r'? | |----------|--------|------------| | 1 | s | No | | 2 | t | No | | 3 | **r** | ✓ (1st) | | 4 | a | No | | 5 | w | No | | 6 | b | No | | 7 | e | No | | 8 | **r** | ✓ (2nd) | | 9 | **r** | ✓ (3rd) | | 10 | y | No | ## Answer: **3** The word "strawberry" contains **three 'r's** — one at position 3, and two consecutive 'r's at positions 8 and 9.

25.6s response455 output tokens

Debug This Error

I'm getting the following error in my Node.js application: TypeError: Cannot read properties of undefined (reading 'map') at UserList (/app/components/UserList.js:12:25) at renderWithHooks (/app/node_modules/rea…

# Error Analysis The error `Cannot read properties of undefined (reading 'map')` tells you that **`users` is `undefined`** when your component runs. JavaScript can't call `.map()` on `undefined`, so it throws this error. The component itself is fine syntactically — the problem is that the `users` prop isn't being passed, or hasn't loaded yet. ## Common Causes **1. The prop isn't passed at all:** ```jsx <UserList /> // No users prop → users is undefined ``` **2. Data is still loading (async fetch):** ```jsx const [users, setUsers] = useState(); // undefined initially ``` **3. API returns undefined:** ```jsx const { data } = await fetch('/api/users').then(r => r.json()); setUsers(data);…

32.8s response710 output tokens

Code Review

Please review the following Python function and suggest improvements for readability, performance, and best practices: def get_data(url, retries=3): import requests import time for i in range(retries):…

## Code Review This function works for the happy path, but it has several issues around error handling, retries, and readability. Here's a breakdown. ### 🐛 Bugs & Correctness Issues | # | Issue | Why It Matters | |---|-------|----------------| | 1 | `except:` with no exception type | Swallows `KeyboardInterrupt`, `SystemExit`, and *everything else* (typos, `MemoryError`). You lose visibility into real failures. | | 2 | Retries on every non‑200, including `4xx` | A `404` or `401` will never succeed by retrying — it just wastes time and rate-limit budget. | | 3 | `if item["active"] == True` | Both linters (PEP 8 E712) and Python idiom prefer `if item["active"]:`. | | 4 | `item["active"]` (…

33.4s response3977 output tokens

Compare these answers side by side with other models →

Frequently asked questions

How much does the MiniMax M3 API cost?: On FastMetal, MiniMax M3 is billed per token in yen: ¥53.61 per 1M input tokens and ¥214.44 per 1M output tokens, before tax. Usage is drawn from a prepaid balance; there is no subscription or monthly fee.
Can I call MiniMax M3 with the OpenAI SDK?: Yes. Point base_url at https://api.fastmetal.ai/v1 and pass "minimax-m3" as the model; existing OpenAI-style code works unchanged, including streaming, tool calls and structured output.
What is the context window of MiniMax M3?: 1,048,576 tokens, shared between the prompt and the response.
How does MiniMax M3 rank?: #74 on the public arena's Japanese board (ELO 1,398). Ranks move as the leaderboard is updated.
What do I need to try MiniMax M3?: Create an account and add credit; the model is then available both in the browser chat and over the API. There is no contract or minimum spend.