DataLearner logo
MI

MiniMax M3

Multimodal modelCoding modelMiniMax M3

MiniMax M3

Release date: 2026-06-01Updated: 2026-06-14 21:29:58.4125,192
Parameters
428B
Context length
1M
Chinese support
Supported
Reasoning ability

MiniMax M3 于 2026 年 6 月 1 日正式发布,采用 MoE 架构(428B 总参数,23B 每 token 激活),并引入自研稀疏注意力架构 MSA(MiniMax Sparse Attention)。在 1M token 上下文下解码速度较上代提升 15.6 倍,支持最高 1M tokens 超长上下文与原生多模态(图片、视频输入及桌面操作)。SWE-Bench Pro 达到 59.0%(超越 GPT-5.5 和 Gemini 3.1 Pro),BrowseComp 得分 83.5(超越 Opus 4.7)。模型权重已在 HuggingFace 和 GitHub 开源。

Data sourced primarily from official releases (GitHub, Hugging Face, papers), then benchmark leaderboards, then third-party evaluators. Learn about our data methodology

MiniMax M3

Model basics

Reasoning traces
Supported
Thinking modes
Thinking Mode (Default)Standard Mode
Context length
1M tokens
Max output length
512K tokens
Model type
Multimodal model
Modality (in / out)
Text, Image, Video → Text
Release date
2026-06-01
Model file size
No data
MoE architecture
Yes
Total params / Active params
428B / 23B
Knowledge cutoff
No data
MiniMax M3

Open source & experience

MiniMax M3

Official resources

MiniMax M3

API details

API speed
3/5
💡Default unit: $/1M tokens. If vendors use other units, follow their published pricing.
Standard
TypeConditionInputOutput
Text-¥2.10/ 1M¥8.40/ 1M
TextContext <= 524288¥4.20/ 1M¥16.80/ 1M
Turbo
TypeConditionInputOutput
Text-¥3.15/ 1M¥12.60/ 1M
TextContext <= 524288¥6.30/ 1M¥25.20/ 1M
Cache PricingPrompt Cache
TypeTTLWriteRead
Text--¥0.420/ 1M
MiniMax M3

Benchmark Results

MiniMax M3 currently shows benchmark results led by BrowseComp (10 / 50, score 83.50), SWE-Bench Pro - Public (10 / 49, score 59), LiveBench (40 / 115, score 70.02). This page also consolidates core specs, context limits, and API pricing so you can evaluate the model from benchmark results and deployment constraints together.

Thinking
Tool usage
Internet

Coding and Software Engineer

1 evaluations
Benchmark / mode
Score
Rank/total
SWE-Bench Pro - Public
Thinking ModeTools
59
10 / 49

AI Agent - Information Search

1 evaluations
Benchmark / mode
Score
Rank/total
BrowseComp
Thinking ModeToolsInternet
83.50
10 / 50

General Knowledge

1 evaluations
Benchmark / mode
Score
Rank/total
70.02
40 / 115

AI Agent - Tool Usage

2 evaluations
Benchmark / mode
Score
Rank/total
OSWorld-Verified
Thinking ModeTools
70
14 / 20
TerminalBench 2.1
Thinking ModeTools
66
18 / 22

Compare with other models

No curated comparisons for this model yet.

Want a custom combination? Open the compare tool

MiniMax M3

Publisher

MiniMax M3

Model Overview

MiniMax M3 于 2026 年 6 月 1 日正式发布,采用 MoE 架构(428B 总参数,23B 每 token 激活),并引入自研稀疏注意力架构 MSA(MiniMax Sparse Attention)。在 1M token 上下文下解码速度较上代提升 15.6 倍,支持最高 1M tokens 超长上下文与原生多模态(图片、视频输入及桌面操作)。SWE-Bench Pro 达到 59.0%(超越 GPT-5.5 和 Gemini 3.1 Pro),BrowseComp 得分 83.5(超越 Opus 4.7)。模型权重已在 HuggingFace 和 GitHub 开源。

DataLearner on WeChat

Follow DataLearner on WeChat for AI model updates and research notes.

DataLearner WeChat QR code