Enter opens the highlighted result. Esc closes.

MiMo-V2.5 vs MiMo-V2.5-Pro

MiMo-V2.5 leads MiMo-V2.5-Pro on coding, 50.0 to 31.6 on the Vibe Coding Index, and MiMo-V2.5 costs 2.7x less per output token ($0.336 vs $0.9152 per million). Updated October 22, 2026; 67 benchmarks have a result for both.

Replace MiMo-V2.5 Replace MiMo-V2.5-Pro
MiMo-V2.5Xiaomi
Coding
50.0
Overall
50.7
Output $/M
$0.336
5
MiMo-V2.5-ProXiaomi
Coding
31.6
Overall
39.3
Output $/M
$0.9152
62
vs boards won of 67 shared

MiMo-V2.5 or MiMo-V2.5-Pro: which to pick

Pick MiMo-V2.5 for coding: it scores 50.0 on the Vibe Coding Index against 31.6 for MiMo-V2.5-Pro, though on 3 benchmarks against 4.

Pick MiMo-V2.5-Pro for work measured by LMArena WebDev or AA Coding Index, where it leads.

Model AMiMo-V2.5Xiaomi50.0/1003 independent benchmarks$0.01 per solved taskModel BMiMo-V2.5-ProXiaomi31.6/1004 independent benchmarks$0.03 per solved task
Shared results
67
MiMo-V2.5 only
13
MiMo-V2.5-Pro only
3
Like-for-like categories
8 of 9

Only benchmarks that tested both models count toward a pick; a result only one model has is shown but not compared.

Which one for your work

Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.

At a glance

Coding index
MiMo-V2.5 50.0 from 3 benchmarks, MiMo-V2.5-Pro 31.6 from 4 benchmarks.
Widest category
General, 8.2 points apart on average over 20 shared benchmarks.
Widest single gap
LMArena Text: Japanese: MiMo-V2.5 1306 against MiMo-V2.5-Pro 1410, 17.3 points apart on a 0 to 100 scale.
Output price
MiMo-V2.5 $0.336 against MiMo-V2.5-Pro $0.9152 per million output tokens, 2.7x apart.
Context window
Both take 1.05M tokens.

Benchmark by benchmark

First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.

Boards won, where both have a result
CategoryMiMo-V2.5BarsMiMo-V2.5-ProTies
Coding311
Intelligence01
Agentic work06
Reasoning06
Math03
Knowledge18
Web design17
General020
All shared boards562
CodingLike-for-likeSix of these boards feed the Vibe Coding Index311
Coding benchmarks for MiMo-V2.5 and MiMo-V2.5-Pro
BenchmarkMiMo-V2.5MiMo-V2.5-ProLeader
LMArena WebDev14381480MiMo-V2.5-Pro
AA Coding Index56.860.2MiMo-V2.5-Pro
LiveCodeBench (Vals AI)81.5%81.3%MiMo-V2.5
SWE-bench Verified (Vals AI)71.0%74.0%MiMo-V2.5-Pro
Terminal-Bench 4 (Vals AI)Not measured0.5%Only MiMo-V2.5-Pro measured
Vibe Code Bench (Vals AI)42.2%34.1%MiMo-V2.5
Code migration (Vals AI)14.2%21.6%MiMo-V2.5-Pro
Terminal-Bench 2.1 (Vals AI)60.7%57.3%MiMo-V2.5
SciCode43.9%50.6%MiMo-V2.5-Pro
Terminal-Bench Hard (Artificial Analysis run)41.7%43.2%MiMo-V2.5-Pro
ALE-Bench514.0899.8MiMo-V2.5-Pro
LMArena Text: coding14691501MiMo-V2.5-Pro
LMArena Text: software and IT14631495MiMo-V2.5-Pro
LMArena WebDev: React14251469MiMo-V2.5-Pro
LMArena WebDev: HTML14291473MiMo-V2.5-Pro
Neither model has a result on the other 24 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index01
Intelligence benchmarks for MiMo-V2.5 and MiMo-V2.5-Pro
BenchmarkMiMo-V2.5MiMo-V2.5-ProLeader
AA Intelligence Index21.826.0MiMo-V2.5-Pro
Neither model has a result on the other 1 board here
Agentic workLike-for-likePublic boards, shown as published, kept out of the index06
Agentic work benchmarks for MiMo-V2.5 and MiMo-V2.5-Pro
BenchmarkMiMo-V2.5MiMo-V2.5-ProLeader
AA Agentic Index15.821.3MiMo-V2.5-Pro
Finance Agent (Vals AI)36.7%41.5%MiMo-V2.5-Pro
Tax Agent Bench (Vals AI)7.8%11.5%MiMo-V2.5-Pro
Legal Research Bench (Vals AI)9.1%15.9%MiMo-V2.5-Pro
Excel Modeling Benchmark (Vals AI)55.1%55.2%MiMo-V2.5-Pro
τ²-Bench Telecom90.6%94.2%MiMo-V2.5-Pro
Neither model has a result on the other 35 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index06
Reasoning benchmarks for MiMo-V2.5 and MiMo-V2.5-Pro
BenchmarkMiMo-V2.5MiMo-V2.5-ProLeader
GPQA Diamond81.6%82.6%MiMo-V2.5-Pro
MMMU Pro (Vals AI)80.0%Not measuredOnly MiMo-V2.5 measured
GPQA Diamond (Artificial Analysis run)84.9%86.6%MiMo-V2.5-Pro
Humanity's Last Exam (Artificial Analysis run)27.2%35.7%MiMo-V2.5-Pro
LMArena Text: hard prompts14501486MiMo-V2.5-Pro
LMArena Text: expert14591501MiMo-V2.5-Pro
LMArena Text: hard prompts in English14611494MiMo-V2.5-Pro
LMArena Vision: diagrams1269Not measuredOnly MiMo-V2.5 measured
LMArena Vision: homework1262Not measuredOnly MiMo-V2.5 measured
Neither model has a result on the other 13 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index03
Math benchmarks for MiMo-V2.5 and MiMo-V2.5-Pro
BenchmarkMiMo-V2.5MiMo-V2.5-ProLeader
ProofBench16.0%22.0%MiMo-V2.5-Pro
LMArena Text: math14361481MiMo-V2.5-Pro
LMArena Text: mathematical work14511491MiMo-V2.5-Pro
Neither model has a result on the other 10 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index18
Knowledge benchmarks for MiMo-V2.5 and MiMo-V2.5-Pro
BenchmarkMiMo-V2.5MiMo-V2.5-ProLeader
MMLU Pro82.9%84.6%MiMo-V2.5-Pro
LegalBench (Vals AI)78.9%77.1%MiMo-V2.5
CorpFin (Vals AI)59.9%61.4%MiMo-V2.5-Pro
TaxEval (Vals AI)71.8%73.8%MiMo-V2.5-Pro
MortgageTax (Vals AI)59.3%Not measuredOnly MiMo-V2.5 measured
MedCode (Vals AI)31.9%32.5%MiMo-V2.5-Pro
LMArena Text: science14401482MiMo-V2.5-Pro
LMArena Text: medicine and healthcare14321473MiMo-V2.5-Pro
LMArena Text: legal and government14251474MiMo-V2.5-Pro
LMArena Text: business and finance14321466MiMo-V2.5-Pro
Neither model has a result on the other 8 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index17
Web design benchmarks for MiMo-V2.5 and MiMo-V2.5-Pro
BenchmarkMiMo-V2.5MiMo-V2.5-ProLeader
Design Arena: websites12651266MiMo-V2.5-Pro
Design Arena: UI components12551246MiMo-V2.5
Design Arena: game dev12491261MiMo-V2.5-Pro
Design Arena: data viz12501257MiMo-V2.5-Pro
Design Arena: 3D12151243MiMo-V2.5-Pro
Design Arena: all code categories12601265MiMo-V2.5-Pro
Design Arena: SVG11791181MiMo-V2.5-Pro
Design Arena: ASCII art11401156MiMo-V2.5-Pro
Neither model has a result on the other 1 board here
GeneralLike-for-likePublic boards, shown as published, kept out of the index020
General benchmarks for MiMo-V2.5 and MiMo-V2.5-Pro
BenchmarkMiMo-V2.5MiMo-V2.5-ProLeader
SAGE (Vals AI)43.3%Not measuredOnly MiMo-V2.5 measured
MedScribe (Vals AI)72.2%83.7%MiMo-V2.5-Pro
Vals Multimodal Index52.8%Not measuredOnly MiMo-V2.5 measured
IFBench67.1%79.9%MiMo-V2.5-Pro
Long context reasoning (AA-LCR)73.0%79.7%MiMo-V2.5-Pro
DTBenchNot measured84.5%Only MiMo-V2.5-Pro measured
LMCANot measured29.5%Only MiMo-V2.5-Pro measured
LMArena Text14281464MiMo-V2.5-Pro
LMArena Text: instruction following14351475MiMo-V2.5-Pro
LMArena Text: creative writing13931438MiMo-V2.5-Pro
LMArena Vision1247Not measuredOnly MiMo-V2.5 measured
LMArena Text: English14471477MiMo-V2.5-Pro
LMArena Text: non-English14051448MiMo-V2.5-Pro
LMArena Text: Chinese14691507MiMo-V2.5-Pro
LMArena Text: German14211459MiMo-V2.5-Pro
LMArena Text: Japanese13061410MiMo-V2.5-Pro
LMArena Text: Korean13611435MiMo-V2.5-Pro
LMArena Text: French14481487MiMo-V2.5-Pro
LMArena Text: Spanish14141469MiMo-V2.5-Pro
LMArena Text: Russian13961449MiMo-V2.5-Pro
LMArena Text: Polish14131464MiMo-V2.5-Pro
LMArena Text: multi-turn14451476MiMo-V2.5-Pro
LMArena Text: longer queries14451481MiMo-V2.5-Pro
LMArena Text: writing and language14091451MiMo-V2.5-Pro
LMArena Text: entertainment and media13991438MiMo-V2.5-Pro
LMArena Vision: English1259Not measuredOnly MiMo-V2.5 measured
LMArena Vision: Chinese1326Not measuredOnly MiMo-V2.5 measured
LMArena Vision: reading text1256Not measuredOnly MiMo-V2.5 measured
LMArena Vision: humor1226Not measuredOnly MiMo-V2.5 measured
LMArena Vision: recognizing things1189Not measuredOnly MiMo-V2.5 measured
LMArena Vision: creative writing1230Not measuredOnly MiMo-V2.5 measured
Neither model has a result on the other 15 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for MiMo-V2.5 and MiMo-V2.5-Pro
BenchmarkMiMo-V2.5MiMo-V2.5-ProLeader
Neither model has a result on 5 boards here

What each workload costs

Three fixed token mixes at each model's listed price, so per-token rates become a bill.

Chat turn

1K input + 500 output tokens

MiMo-V2.5$0.0003Fits in one request
MiMo-V2.5-Pro$0.0009Fits in one request

MiMo-V2.5 costs less

Repository review

50K input + 3K output tokens

MiMo-V2.5$0.0094Fits in one request
MiMo-V2.5-Pro$0.026Fits in one request

MiMo-V2.5 costs less

Agent loop

200K input + 10K output tokens

MiMo-V2.5$0.037Fits in one request
MiMo-V2.5-Pro$0.101Fits in one request

MiMo-V2.5 costs less

Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.

Specification differences

The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.

Specifications of MiMo-V2.5 and MiMo-V2.5-Pro
SpecificationXiaomi MiMo-V2.5Xiaomi MiMo-V2.5-Pro
MakerXiaomiXiaomi
WeightsOpen weightsOpen weights
ReleasedApril 22, 2026April 22, 2026
Context window1.05M1.05M
Input $/Mtok$0.168$0.4575
Output $/Mtok$0.336$0.9152
Coding index50.031.6
Independent measurements behind it34
Cost per solved task$0.01$0.03
SourceOpenRouter listingOpenRouter listing

Questions

Is MiMo-V2.5 better than MiMo-V2.5-Pro for coding?

Pick MiMo-V2.5 for coding: it scores 50.0 on the Vibe Coding Index against 31.6 for MiMo-V2.5-Pro, though on 3 benchmarks against 4. Pick MiMo-V2.5-Pro for work measured by LMArena WebDev or AA Coding Index, where it leads.

Which is cheaper, MiMo-V2.5 or MiMo-V2.5-Pro?

MiMo-V2.5 costs $0.168 per million input tokens and $0.336 per million output tokens; MiMo-V2.5-Pro costs $0.4575 and $0.9152. Per solved SWE-bench Verified task, MiMo-V2.5 costs $0.01 and MiMo-V2.5-Pro $0.03.

Which has the bigger context window, MiMo-V2.5 or MiMo-V2.5-Pro?

Neither: both take 1.05M tokens.

How many benchmarks is the MiMo-V2.5 vs MiMo-V2.5-Pro comparison based on?

67 benchmarks have a result for both models as of October 22, 2026. MiMo-V2.5's coding index rests on 3 independent benchmarks and MiMo-V2.5-Pro's on 4. Where only one model has a result, the page says so instead of naming a winner.

Compare next

Embed this comparison on your site

Free under CC BY 4.0, while the source link stays visible. Preview the frame · Terms and other embeds

<iframe src="https://llmbenchmark.ai/embed/compare/xiaomi-mimo-v2-5-vs-xiaomi-mimo-v2-5-pro" title="MiMo-V2.5 vs MiMo-V2.5-Pro, LLMBenchmark.ai" width="100%" height="360" style="border:0;max-width:100%" loading="lazy"></iframe>
<p style="font:13px/1.4 sans-serif;margin:6px 0 0">Source: <a href="https://llmbenchmark.ai/compare/xiaomi-mimo-v2-5-vs-xiaomi-mimo-v2-5-pro">LLMBenchmark.ai, MiMo-V2.5 vs MiMo-V2.5-Pro</a></p>