跳转到主要内容

MiniMax-M2.5

MiniMaxMiniMaxOpen WeightMIT · Commercial OK

描述

MiniMax M2.5 is the world's first production-level model designed natively for Agent scenarios. Building on the M2.1 foundation, M2.5 delivers significant improvements in programming, tool calling, search, and office productivity. With only 10B activation parameters from its 230B MoE architecture, it achieves competitive performance against top international models like Claude Opus 4.6 while maintaining high throughput and efficient inference. M2.5 supports full-stack development for PC, App, and cross-platform applications, and excels in agentic workflows including automated customer support, data-analysis pipelines, and complex task execution.

发布日期
2026-02-12
参数规模
230.0B
上下文长度
197K
支持模态
image, text

能力雷达图

37
general
38
coding
85
reasoning
57
science估算
70
agents
60
multimodal

Science 在缺少专门科学评测时使用推理能力代理估算。

排行榜排名

领域#排名分数来源
智能体与工具27
65.0
LS
代码能力榜68
68.0
AA
通用能力榜46
79.0
AA
科学能力71
67.0
AA

基准测试分数 (LLM Stats)

Agents

BrowseComp76.3%自报
MEWC74.4%自报
SWE-Bench Pro55.4%自报
VIBE-Pro54.2%自报

Code

SWE-Bench Verified80.2%自报
Multi-SWE-Bench51.3%自报

Finance

GDPval-MM59.0%自报

General

BFCL_v3_MultiTurn76.8%自报

AA 评测指数

Intelligence Index
41.9
Coding Index
37.4
Tau2
1.0
Gpqa
0.8
Ifbench
0.7
Lcr
0.7
Scicode
0.4
Terminalbench Hard
0.3
Hle
0.2

LLM Stats 分类评分

Frontend Development
80
Search
80
Agents
70
Code
60
Finance
60
General
60
Multimodal
60
Reasoning
60

定价

输入价格$0.3 / 1M tokens
输出价格$1.2 / 1M tokens
混合价格(3:1)$0.525 / 1M tokens

速度

Tokens/秒84.0 tokens/s
首Token延迟1.44s
首回答延迟25.25s

可用提供商

(LS 内部计价单位)
提供商输入价格输出价格
MiniMax300K1.2M

外部链接