Noometry

Xiaomi, open weights

MiMo-V2.5-Pro

MiMo-V2.5-Pro by Xiaomi ranks 74th of 354 ranked models on the Noometry Index as of October 2026, with a score of 45.2. Its strongest category is instruction following, where it ranks 21st. API pricing starts at $0.43 per million input tokens and $0.87 per million output tokens, with a 1.05M-token context window.

Last verified

Noometry rank
#74 of 354
Index score
45.2
Evidence
Confirmed 27 results
Provider
Xiaomi
Released
April 22, 2026
Weights
Open weights
Reasoning
Yes
Context window
1.05M
Max output
131K
Input price
$0.43 / M
Output price
$0.87 / M
Blended price
$0.54 / M
Output speed
Not measured
Value
#74 of 219
Knowledge cutoff
December 2024
Input
text
Hugging Face
XiaomiMiMo/MiMo-V2.5-Pro

Category scores

Each category score combines every public result we have in that category.

  1. Coding 47.4
  2. Reasoning 26.8
  3. Math 40.0
  4. Knowledge 42.2
  5. Multilingual 55.1
  6. Instruction Following 77.5
  7. Long Context 45.4
  8. Writing & Preference 65.3
CategoryScoreRankResults
Coding47.4#603
Reasoning26.8#1305
Math40.0#962
Knowledge42.2#981
Multilingual55.1#341
Instruction Following77.5#211
Long Context45.4#371
Writing & Preference65.3#495

Strengths and weaknesses

Categories where MiMo-V2.5-Pro places highest and lowest among the models ranked in each, with its score against that category's median.

Strongest categories

CategoryScorevs medianRank
Instruction Following77.5+6.2#21 of 305, top 7%
Multilingual55.1+7.7#34 of 297, top 12%
Long Context45.4+4.5#37 of 296, top 13%

Weakest categories

CategoryScorevs medianRank
Reasoning26.8+3.2#130 of 350, top 38%
Knowledge42.2+4.9#98 of 314, top 32%
Math40.0+3.5#96 of 327, top 30%

Closest competitors

The models ranked just above and below MiMo-V2.5-Pro. When scores are this close, price and speed are often the better way to choose.

ModelRankScoreBlended $/MSpeed
Kimi K2 Thinking Turbo#7045.8——Compare
Qwen3.5 Max Preview#7145.3——Compare
Qwen3.7 Plus#7245.3$0.70—Compare
Hy4 preview#7345.3$1.13—Compare
Gemini 2.5 Pro#7545.0$3.445Compare
GPT-5.4 mini#7645.0$1.6910Compare
Amazon Nova Experimental Chat 26 02 10#7744.5——Compare
DeepSeek-V3.2-Exp#7844.3$0.2916Compare

Sponsored placements are available on pages like this one. Advertise on Noometry

Benchmark results

Every published result we track, with its source. Bold rows are the ones used for ranking; where several exist we prefer independent runs over self-reported numbers.

Coding

BenchmarkScorePositionSettingSourceDate
LMArena WebDev1479#52 of 113, top 47%LMArena2026-10-08
SciCode50.2%#43 of 121, top 36%Epoch AI
LMArena Coding1503#18 of 294, top 7%LMArena2026-10-08
ALE-Bench899.8#51 of 105, top 49%Epoch AI

Reasoning

BenchmarkScorePositionSettingSourceDate
NYT Connections (extended)34.4%#71 of 91, top 79%Lech Mazur benchmarks
CritPt4%#60 of 134, top 45%Epoch AI
LMArena Hard Prompts1488#19 of 297, top 7%LMArena2026-10-08
DTBench84.5%#55 of 151, top 37%Epoch AI
LMCA29.5%#74 of 125, top 60%Epoch AI

Math

BenchmarkScorePositionSettingSourceDate
ProofBench22%#46 of 77, top 60%Epoch AI
LMArena Math1481#28 of 285, top 10%LMArena2026-10-08

Knowledge

BenchmarkScorePositionSettingSourceDate
LMArena Expert1503#22 of 273, top 9%LMArena2026-10-08

Multilingual

BenchmarkScorePositionSettingSourceDate
LMArena Non-English1449#34 of 297, top 12%LMArena2026-10-08
LMArena Chinese1507#41 of 285, top 15%LMArena2026-10-08
LMArena French1488#21 of 223, top 10%LMArena2026-10-08
LMArena German1458#36 of 231, top 16%LMArena2026-10-08
LMArena Japanese1412#49 of 211, top 24%LMArena2026-10-08
LMArena Korean1437#26 of 213, top 13%LMArena2026-10-08
LMArena Russian1450#44 of 283, top 16%LMArena2026-10-08
LMArena Spanish1471#23 of 226, top 11%LMArena2026-10-08

Instruction Following

BenchmarkScorePositionSettingSourceDate
LMArena Instruction Following1477#18 of 298, top 7%LMArena2026-10-08

Long Context

BenchmarkScorePositionSettingSourceDate
LMArena Longer Query1483#18 of 291, top 7%LMArena2026-10-08

Writing & Preference

BenchmarkScorePositionSettingSourceDate
LMArena Text1465#31 of 297, top 11%LMArena2026-10-08
LMArena Creative Writing1440#37 of 295, top 13%LMArena2026-10-08
EQ-Bench Creative Writing1493#56 of 115, top 49%EQ-Bench
EQ-Bench 41208#15 of 28, top 54%EQ-Bench
LMArena Multi-Turn1477#23 of 295, top 8%LMArena2026-10-08

API pricing by provider

RouteInput $/MOutput $/MCached input $/MChecked
deepinfra$1$3$0.202026-10-10
openrouter$0.43$0.87$0.00362026-10-10
xiaomi$0.43$0.87$0.00362026-10-10

All Xiaomi API prices → Estimate your cost →

Compare MiMo-V2.5-Pro

Other Xiaomi models

Frequently asked questions

How good is MiMo-V2.5-Pro?

MiMo-V2.5-Pro by Xiaomi ranks 74th of 354 ranked models on the Noometry Index as of October 2026, with a score of 45.2. Its strongest category is instruction following, where it ranks 21st. API pricing starts at $0.43 per million input tokens and $0.87 per million output tokens, with a 1.05M-token context window.

How much does MiMo-V2.5-Pro cost?

MiMo-V2.5-Pro costs $0.43 per million input tokens and $0.87 per million output tokens on Xiaomi's own API, with cached input at $0.0036.

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

MiMo-V2.5-Pro accepts up to 1.05M tokens of input and can write up to 131K tokens in one response.

Is MiMo-V2.5-Pro open source?

Yes. MiMo-V2.5-Pro's weights are downloadable from Hugging Face (XiaomiMiMo/MiMo-V2.5-Pro); check the license for commercial terms.

What are MiMo-V2.5-Pro's strengths and weaknesses?

Relative to other ranked models, MiMo-V2.5-Pro places best in instruction following, multilingual, long context and lowest in reasoning, knowledge, math.

What is MiMo-V2.5-Pro best at?

Its best category is instruction following, where it ranks 21st on Noometry.