Image preference Cap is the peer rank for this modality — preference arenas and measured boards (VBench / Open ASR / AA categories) when they exist. Empty axes stay blank; Cap never copies an overall score into realism or motion.
Analysis Media Arena charts (AA Text-to-Image, LMArena I2V, TTS Arena, …) feed this portfolio when the catalog worker has matched this slug. Cards below list Cap axes, arena scores, and evidence sources — not the Language Model Matrix.
IMAGE PREFERENCE CAP · QUALITY AXES
Human preference
93.8
CADLE INSIGHTS
Only measured signals
IMAGE CAP
90
#6/122 · 58% confidence
HUMAN PREFERENCE
93.8
Media Cap quality axis
STRONGEST SIGNAL
Human preference
Cap axis 93.8
BLENDED $/1M
$43.00
In $5.00 · Out $38.00
VALUE INDEX
2.1
Quality ÷ blended $/1M (higher is better)
SPEED · COST · QUALITY
Measured matrix / AA / gateway rows only.
Input $/1M
$5.00
Output $/1M
$38.00
DESCRIPTION + DETAILS
PROVIDER
Microsoft
STATUS
ACTIVE
COMMUNITY REVIEWS
OPERATOR NOTES
Quality, latency, cost, and failure modes — not marketing copy.
No reviews yet
WRITE A REVIEW
Tap a star to rate
Stored on this device · synced when signed in
No operator notes yet
First review sets the tone for how teams evaluate MAI-Image-2.6.
MODEL COMPARISON
Metric
MAI-Image-2.6
Provider
Microsoft
Release
Input / 1M
$5.00
Output / 1M
$38.00
Context
No data yet
Cadle Cap
90
Cap board
#6/122
Human preference
93.8
DATA PROVENANCE
Scores may use different scales — check the Benchmarks tab for source links.
Supabase models
Cadle public observations
Open benchmark data
LMArena Text to image: #5 · Elo 1332 · 25,091 votes