XiaomiChina· released Sep 2026· open weights

MiMo V2.6 Pro

Name: MiMo V2.6 Pro
Author: Xiaomi

Provisional: 7 comparable tests so far. Ranked once it reaches 6 across 2 counting categories.

Xiaomi, China. One of 4 models from this lab on the Index.

Third-party publishedLow confidence (thin coverage)

Snapshot November 10, 2026

OpenCharts Index

74.6

Provisional: not enough comparable tests or counting categories to rank yet · 7 comparable tests · 0 counting categories

Drop any single test and the Index lands between 71.2 and 78.1.

Needs more results (0 of 5 so far)

Distance to today's frontier on the hardest tests. Not a measure of intelligence.

What it's known for

MiMo V2.6 Pro, in 7 results.

Its widest gap is on Code Arena (WebDev), 50.6% behind the best published result. No comparable result yet in knowledge. Between 25% and 30% behind the best published results, on average.

Strongest category

No category counts yet: a category counts once it has more than one comparable test and the model has taken at least half of them. The scores it has are below.

Best single result

SciCode

Write research-grade scientific code, one sub-problem at a time, that passes unit tests.

60.9%

91.0% of the best (66.9%) · #7 of 96.

Scientific programming needs the math and the code to be right at the same time.

Where it trails

Code Arena (WebDev)

Build a web app from the same prompt as a rival model; real users vote blind on the result.

1,624

50.6 points short of parity with the board leader (1,818) · #23 of 110.

People judging finished apps side by side is the most honest measure of front-end quality.

Not measured yet

No comparable result yet in this category. A blank is a blank, never a zero, and it does not lower the Index.

Knowledge

Covered, but not averaged into the Index:

Reasoning1 of 8 tests · needs 4 · mean 73.8
Coding2 of 6 tests · needs 3 · mean 70.2
Math1 of 4 tests · needs 2 · mean 70.0
Agentic1 of 6 tests · needs 3 · mean 64.6
Multimodal1 of 4 tests · needs 2 · mean 82.3
Human preferencesingle-test category, shown beside the Index · 86.9

Every score, with its source

See it at work

What MiMo V2.6 Pro was asked to do, and how it did.

Every test it has taken, by category, with the kind of task it faced, who produced the number and where the result landed against the best published one. The examples are original and illustrative, never items from the datasets themselves.

Human preference · mean 86.9 across 1 comparable test · a single-test category, shown beside the Index and never averaged into it

Text ArenaLeaderboard

Answer the same prompt as a rival model; real users vote blind on which answer they prefer.

1,480

43% expected win rate against the board leader (1,526)
#27 of 113

board 2026-11-09

Source: Arena

Every result is read against the best published one on its test: percent scores as a share of the best above chance, Arena ratings as the expected win rate against the board leader, open-ended values on a log scale. Sitting a harder exam never lowers a model. Every number links to the publisher that produced it.

Head to head

MiMo V2.6 Pro against whoever you pick.

Choose a rival. Every counted test both have taken appears side by side as a share of the best published result on that test, with the category means above. A win is a higher share; a gap under 2 points is a tie, because one item on a 45-problem exam is 2.2 points and publishers' own standard errors are of that order.

MiMo V2.6 Pro

provisional·Index 74.6B-

1 : 5

wins · 1 tie · 7 shared

GPT-6 Astra

#1·Index 93.7A

GPT-6 Astra comes out ahead on 5 of the 7 tests both have taken.

Category means

Reasoning
73.8
96.1
Knowledge
—
80.4
Coding
70.2
93.4
Math
70.0
98.8
Agentic
64.6
90.6
Multimodal
82.3
89.6
Human preference
86.9
85.2

Shared tests · biggest gaps first

Code Arena (WebDev)Coding
1,624
1,767
−36.1
ProofBenchMath
70.0%
99.0%
−29.0
CritPtReasoning
26.6%
31.7%
−14.3
APEX-AgentsAgentic
54.4%
64.7%
−12.2
Vision ArenaMultimodal
1,247
1,283
−10.2
SciCodeCoding
60.9%
56.5%
+6.6
Text ArenaHuman preference
1,480
1,474
tie

Bars are distances to the best published result on each test (percent scores above chance, Arena ratings as win rate against the leader, open-ended values on a log scale). Raw values as the publishers report them.

Open GPT-6 Astra

Scores

Every test, every source

The best published run per benchmark, the raw value as the publisher reports it, who produced the number, how close it comes to the best published result on that test, and where the model sits among every model scored on it. Percent scores are read above chance, Arena ratings as a win rate against the board leader, open-ended values on a log scale. A test with too few models to compare against is shown but not counted, and so is an archived test: one whose publisher no longer covers new models.

Category means

Reasoning73.8 · 1/8
Knowledge—
Coding70.2 · 2/6
Math70.0 · 1/4
Agentic64.6 · 1/6
Multimodal82.3 · 1/4
Human preferencebeside86.9 · 1/1
Long context—

Mean · tests taken of the category's comparable tests. A faded category either rests on fewer than half of them or is a single-test category (marked “beside”), which is shown next to the Index and never averaged into it.

CritPtLeaderboardReasoning

26.6%

73.8% of the best (36.0%)
83th percentile of 102

Source: Epoch AI — AI Benchmarking Hub

SciCodeLeaderboardCoding

60.9%

91.0% of the best (66.9%)
94th percentile of 96 · n = 338

Source: Epoch AI — AI Benchmarking Hub

Code Arena (WebDev)LeaderboardCoding

1,624

25% expected win rate against the board leader (1,818)
80th percentile of 110

board 2026-11-09