The AI arena is free today

Open Superagent
MiniMaxReleased on Jun 16, 2025

MiniMax M1 40K: API Pricing, Context Window & Benchmarks

MiniMax M1 40K is a language model from MiniMax, released in June 2025.

MiniMax-M1 is an open-source, large-scale reasoning model that uses a hybrid-attention architecture for efficient long-context processing. It supports up to a 1 million token context window and 80,000-token reasoning output, matching

MiniMax M1 40K benchmarks

Capability tiers

Standing within each category, adjusted for leaderboard depth.

Real tasks performance

How MiniMax M1 40K performs across real-world prompt categories.

Performance by conversation depth

How MiniMax M1 40K holds up as conversations get longer.

Quality Tracker

MiniMax M1 40K Performance Across Datasets

Scores sourced from the model's scorecard, paper, or official blog posts

LLM Stats Logollm-stats.com - Mon Aug 24 2026
Notice missing or incorrect data?

MiniMax M1 40K model size

MiniMax M1 40K has 456 billion parameters and was trained on 7.5 trillion tokens. See how it compares to other models in the same parameter range.

ParametersTraining tokens
456B
7.5Ttokens
16× tokens-to-params ratio
Frontier (200B+)
456B
1B7B70B405B

MiniMax M1 40K API

Available from the model provider

MiniMax M1 40K has an official provider API. It is not currently routed through the LLM Stats gateway.

Read the official API documentation

MiniMax M1 40K latency

MiniMax M1 40K time to first token, sustained output throughput, and failed-request rate from live API traffic over the trailing 7 days.

MiniMax M1 40K examples

Recent arena outputs from MiniMax M1 40K, picked from the highest-ranked matchups.

MiniMax M1 40K license

MiniMax M1 40K is released under the MIT license, which permits commercial use, has 456.0B parameters.

License
MIT
Commercial use allowed
Parameters
456.0B

MIT License - allows commercial use

MiniMax M1 40K resources

Official sources for MiniMax M1 40K: api documentation, paper or system card, official launch post, source repository, model weights.

MiniMax M1 40K vs other models

The most-compared alternatives to MiniMax M1 40K are Phi 4 Reasoning Plus, Qwen3 VL 235B A22B Instruct, Nova 2 Omni. Open any pair side-by-side for benchmarks, pricing, context, and latency.

Models like MiniMax M1 40K

Models ranked just above and below MiniMax M1 40K by LLM Stats score.

 

Phi 4 Reasoning Plus

Score pending
 

Qwen3 VL 235B A22B Instruct

Score pending
 

Nova 2 Omni

Score pending
 

Qwen3 VL 32B Instruct

Score pending
 

Qwen3 VL 30B A3B Thinking

Score pending
 

LongCat-Flash-Lite

Score pending

FAQ

Common questions about MiniMax M1 40K.

When was MiniMax M1 40K released?

MiniMax M1 40K was released on June 16, 2025 by MiniMax. This is the official MiniMax M1 40K release date tracked on LLM Stats.

Is MiniMax M1 40K available via API?

Yes, MiniMax M1 40K is available via API. See the official documentation for authentication and endpoint details.

How big is MiniMax M1 40K?

MiniMax M1 40K has 456 billion parameters. It was trained on 7.5 trillion tokens. It ships as an open-weight model, so you can download and run it on your own hardware.

Who created MiniMax M1 40K?

MiniMax M1 40K was created by MiniMax.

What is the license for MiniMax M1 40K?

MiniMax M1 40K is released under the MIT license. This is an open-source / open-weight license that permits self-hosting.

Where is the MiniMax M1 40K paper or technical report?

MiniMax M1 40K has a paper or technical report available at https://arxiv.org/abs/2506.13585. Use that source for architecture, training, release and evaluation details.

What models should I compare MiniMax M1 40K against?

Common MiniMax M1 40K comparisons include MiniMax M1 40K vs Phi 4 Reasoning Plus, MiniMax M1 40K vs Qwen3 VL 235B A22B Instruct, MiniMax M1 40K vs Nova 2 Omni. Compare them side by side for benchmark scores, pricing, context window, latency and API availability.