MiMo-V2-Flash (Non-reasoning)
XiaomiOpen WeightMIT · Commercial OK
描述
MiMo-V2-Flash is a powerful, efficient, and ultra-fast foundation language model that excels in reasoning, coding, and agentic scenarios. It is a Mixture-of-Experts model with 309B total parameters and 15B active parameters, featuring a hybrid attention architecture with sliding-window and full attention (5:1 ratio, 128-token window). Delivers 150 tokens/sec inference with 256k context window.
發布日期
2025-12-16
參數規模
309.0B
上下文長度
262K
支援模態
text
能力雷達圖
40
general
31
coding
67
reasoning
40
science估算
60
agents
0
multimodal
Science 在缺少專門科學評測時使用推理能力代理估算。
排行榜排名
基準測試分數 (LLM Stats)
Agents
Tau-bench
80.3%自報
BrowseComp
58.3%自報
Terminal-Bench 2.0
38.5%自報
Terminal-Bench
30.5%自報
Biology
GPQA
83.7%自報
Code
SWE-Bench Verified
73.4%自報
SWE-bench Multilingual
71.7%自報
Creativity
Arena-Hard v2
86.2%自報
Finance
MMLU-Pro
84.9%自報
General
LiveCodeBench v6
80.6%自報
LongBench v2
60.6%自報
MRCR
45.7%自報
Math
AIME 2025
94.1%自報
HMMT 2025
84.4%自報
Humanity's Last Exam
22.1%自報
AA 評測指數
Math Index67.7
Intelligence Index30.4
Coding Index25.8
Tau20.8
Mmlu Pro0.7
Aime 250.7
Gpqa0.7
Livecodebench0.4
Ifbench0.4
Lcr0.3
Scicode0.3
Terminalbench Hard0.3
Hle0.1
LLM Stats 分類評分
Writing90
Creativity90
Biology80
Chemistry80
Finance80
Healthcare80
Language80
Legal80
Physics80
Frontend Development70
General70
Math70
Reasoning70
Structured Output60
Tool Calling60
Search60
Agents50
Code50
Long Context50
Vision20
定價
輸入價格$0.1 / 1M tokens
輸出價格$0.3 / 1M tokens
混合價格(3:1)$0.15 / 1M tokens
速度
Tokens/秒129.6 tokens/s
首Token延遲1.41s
首回答延遲1.41s
可用提供商
(LS 內部計價單位)暫無提供商資料