MiniMax-M1-80k
456B
Model parameter quantity
MiniMax
Affiliated organization
Open Source
License Type
June 17, 2025
Release time
Model Introduction
The world's first open-weight, large-scale hybrid-attention reasoning model released by Minimax
Comprehensive score
Language dialogue
Knowledge reserve
Reasoning association
Mathematical calculation
Code writing
Command following
Swipe left and right to view more
Language comprehension ability
Capable of understanding complex contexts and generating logically coherent sentences, though occasionally off in tone control.
7.0
Knowledge coverage scope
Possesses core knowledge of mainstream disciplines, but has limited coverage of cutting-edge interdisciplinary fields.
8.2
Reasoning ability
Unable to maintain coherent reasoning chains, often causing inverted causality or miscalculations.
5.2
Model comparison
MiniMax-M1-80k vs Qwen2.5-7B-Instruct
Like Qwen2, the Qwen2.5 language models support up to 128K tokens and can generate up to 8K tokens. They also maintain multilingual support for over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, Arabic, and more.
MiniMax-M1-80k vs GPT-4o-mini-20240718
GPT-4o-mini is an API model produced by OpenAI, with the specific version number being gpt-4o-mini-2024-07-18.
MiniMax-M1-80k vs GPT-4o-mini-20240718
GPT-4o-mini is an API model produced by OpenAI, with the specific version number being gpt-4o-mini-2024-07-18.
MiniMax-M1-80k vs Gemini-2.5-Pro-Preview-05-06
Gemini 2.5 Pro is a model released by Google DeepMind artificial intelligence research team, using version number Gemini-2.5-Pro-Preview-05-06.
MiniMax-M1-80k vs DeepSeek-V2-Chat-0628
DeepSeek-V2 is a strong Mixture-of-Experts (MoE) language model characterized by economical training and efficient inference. It comprises 236B total parameters, of which 21B are activated for each token. Compared with DeepSeek 67B, DeepSeek-V2 achieves stronger performance, and meanwhile saves 42.5% of training costs, reduces the KV cache by 93.3%, and boosts the maximum generation throughput to 5.76 times.
Related model
MiniMax-Text-01
MiniMax-Text-01 is a powerful language model with 456 billion total parameters, of which 45.9 billion are activated per token. To better unlock the long context capabilities of the model, MiniMax-Text-01 adopts a hybrid architecture that combines Lightning Attention, Softmax Attention and Mixture-of-Experts (MoE).
abab6.5
abab6.5 is an API model produced by MiniMax, with the version number being abab6.5. The abab6.5 serie is a trillion-parameter Mixture of Experts (MoE) large language model. The abab6.5 is suitable for complex scenarios, such as application problem calculations, scientific computations, and other similar scenarios. The abab6.5s is suitable for general scenarios.
abab6.5s-chat
abab6.5 is an API model produced by MiniMax, with the version number being abab6.5. The abab6.5 serie is a trillion-parameter Mixture of Experts (MoE) large language model. The abab6.5 is suitable for complex scenarios, such as application problem calculations, scientific computations, and other similar scenarios. The abab6.5s is suitable for general scenarios.
MiniMax-Text-01
MiniMax-Text-01 is a powerful language model with 456 billion total parameters, of which 45.9 billion are activated per token. To better unlock the long context capabilities of the model, MiniMax-Text-01 adopts a hybrid architecture that combines Lightning Attention, Softmax Attention and Mixture-of-Experts (MoE).
abab7-chat-preview
The abab7-preview model, produced by MiniMax, is an API model that shows significant improvements over the abab6.5 series in capabilities such as handling long texts, mathematics, and writing.
Relevant documents
TikTok's Ban Crisis Nears Resolution with Potential New App and Sale
TikTok Sale Nears Completion as New US Version Prepares for LaunchDespite the TikTok divest-or-ban legislation taking effect in January, the platform has maintained US operations with only a brief one-day shutdown. *The Information* now reports that
Amazon Discontinues Shared Prime Free Shipping Benefits Outside Households
Amazon Ends Prime Sharing ProgramAmazon is eliminating the popular feature that allowed Prime members to extend their free shipping benefits to non-household members. According to updated support documentation, this sharing capability will terminate
HMD Scales Back US Operations, Ending Nokia Phone Revival
HMD Global, the Finnish company that revitalized Nokia-branded mobile devices through a licensing agreement over the past decade, has announced a significant reduction in its US market presence. The company appears to have halted all direct sales of
Global Startups Must Navigate AI Policy: Key Strategies to Know
I notice you're asking me to rewrite content that includes an embedded YouTube iframe. However, I'll follow the strict requirements you initially provided:I must preserve all HTML tags exactly as they appear, without modificationI can only rewrite th
Google Leaks Details of Upcoming Android Design Language: Material 3 Expressive
Google Prepares to Unveil Next-Gen Android Design System at I/OGoogle is set to introduce a significant evolution of its Android design language at the upcoming Google I/O developer conference, as revealed through a published event schedule and an ac
