MiMo-V2-Pro

XiaomiProprietary

描述

MiMo-V2-Pro is Xiaomi's flagship foundation model built for real-world agentic workloads. A Mixture-of-Experts model with over 1T total parameters and 42B active, roughly 3x larger than MiMo-V2-Flash. Inherits the Hybrid Attention mechanism with a 7:1 ratio and supports up to 1M-token context. Optimized for coding and agent tasks through post-training scaling across diverse agent scenarios, with native support for complex tool calling and multi-step reasoning.

发布日期

2026-03-18

参数规模

1.0T

上下文长度

1.0M

支持模态

text

能力雷达图

general

coding

reasoning

science

agents

multimodal

排行榜排名

领域	#排名	分数	来源
智能体能力模型榜	94	41.0	LS
代码能力榜	110	71.0	AA
通用能力榜	65	77.0	AA
科学能力	88	72.0	AA

基准测试分数 (LLM Stats)

(LLM Stats (zeroeval))

Agents

DeepSearchQA

86.7%自报

PinchBench

81.0%自报

Claw-Eval

61.5%自报

Terminal-Bench 2.0Stanford × Laude Institute (2026)

57.1%自报

Code

SWE-Bench Verified

78.0%自报

SWE-bench Multilingual

71.7%自报

Communication

Tau2 Telecom

96.8%自报

AA 评测指数

(Artificial Analysis)

Intelligence Index(Artificial Analysis)

41.4

Tau2(Sierra + U Toronto + Vector Institute (2025))

1.0

Gpqa(NYU + Cohere + Anthropic (2023))

0.9

Ifbench(Google Research (2023))

0.7

Lcr(Artificial Analysis)

0.7

Scicode(UIUC + Argonne National Lab (2024))

0.4

Terminalbench Hard(Stanford × Laude Institute (2026))

0.4

Hle(Center for AI Safety + Scale AI (2025))

0.3

LLM Stats 分类评分

(LLM Stats (zeroeval))

Communication

100

Reasoning

Frontend Development

General

Tool Calling

Agents

Code

定价

输入价格免费

输出价格免费

混合价格(3:1)免费

缓存读取价格$0.0036 / 1M tokens

速度

Tokens/秒0.0

首Token延迟0.00s

首回答延迟0.00s

供应商价格排行

3 个供应商

最便宜: Xiaomi最贵: OpenCode Go

供应商输入输出

1Xiaomi最便宜

$0.435

$0.87

2Abacus

3OpenCode Go

比较该模型在不同 API 供应商之间的定价。

外部链接

LLM Stats Artificial Analysis