modelgrep
X

Xiaomi: MiMo-V2.5

xiaomi/mimo-v2.5

16th smartest of 178Cheaper than 76% of paidReasoningToolsJSONVisionAudio
Use via OpenRouter ↗
Intelligence
49.0
16th of 178
Design Elo
1317
UI Component
Speed
55
148th fastest
Latency
2.6s
first token
Input price
$0.140
72nd cheapest
Context
1.0M
131K max out

How it compares

Smarter than91%
of all ranked models
Faster than50%
of all ranked models
Cheaper than76%
of all ranked models

Overview

MiMo-V2.5 is a native omnimodal model by Xiaomi. It delivers Pro-level agentic performance at roughly half the inference cost, while surpassing MiMo-V2-Omni in multimodal perception across image and video understanding...

Benchmarks

independent · via OpenRouter
Artificial Analysis89th percentile
Intelligence Index
49.0
Coding Index
42.1
Agentic Index
65.5
GPQA Diamond
85%
Humanity's Last Exam
25%
SciCode
43%
Tau²-Bench (agentic)
91%
Design Arena · Elo10,550 tournaments
UI Component
1317
codecategories
1305
Website
1305
3D
1303
Game Dev
1299
Data Viz
1281
svg
1208
asciiart
1186

Providers & pricing (1)

ProviderIn $/MOut $/MUptime
Xiaomifp8$0.140$0.280100%

Specifications

Context window1.0M
Max output131K
Knowledge cutoff
Input modalitiestext, audio, image, video
Output modalitiestext
Prompt caching
Cache read price$0.0028/M
ModeratedNo

MiMo-V2.5 FAQ

How much does MiMo-V2.5 cost?

MiMo-V2.5 costs $0.140 per million input tokens and $0.280 per million output tokens via OpenRouter, making it 72nd cheapest of 298 paid models.

How smart is MiMo-V2.5?

MiMo-V2.5 scores 49.0 on the Artificial Analysis Intelligence Index, ranking 16th of 178 benchmarked models, with a GPQA Diamond score of 85%.

How fast is MiMo-V2.5?

MiMo-V2.5 generates around 55 tokens per second with 2.6s time-to-first-token (p50), the 148th fastest tracked model.

What is MiMo-V2.5's context window?

MiMo-V2.5 supports a 1.0M-token context window and can output up to 131K tokens. It accepts text, audio, image, video input.

Compare head-to-head