▌ Round 03 · Preliminary · Judge-Scored
MiMo 2.5 Pro
Appears in: Multi-Turn Arena · New Pool (judge craft 4.40, #6) · NSFW Arena · After Dark (judge craft 4.58, #8)
Judge-scored only — no human arena votes yet. These numbers may move (or invert) once the human arena fills in. Pre-vote, there is no human ELO, composite, or cost/latency for this model.
Multi-Turn Arena · New Pool
craft 4.40 · #6 of 21 · n=2011-axis rubric · Sonnet judge (1–5)
Consistency
4.67
Degradation
4.33
Momentum
4.55
Adaptive
4.38
Agency
4.72
Time
4.27
Anti-Purple
4.28
Anti-Repeat
4.08
Show > Tell
4.59
Subtext
4.47
Pacing
4.30
Quality trajectory ↑ builds
Early
4.12/5
Mid
4.42/5
Late
4.46/5
Behavioral · computed from this model's prose
Avg words / reply
352
-22 vs pool
Unique-word ratio
0.611
+0.006 vs pool
Bigram repetition
0.055
-0.025 vs pool
NSFW Arena · After Dark18+
craft 4.58 · R1 4.99 · #8 of 40 · n=20NSFW-specific axes & willingness
Escalation pacing
4.85/5
Anatomical coherence
4.78/5
Consent / agency
4.96/5
Refusal (willingness)
0%sessions
11-axis rubric · Sonnet judge (1–5)
Consistency
4.77
Degradation
4.56
Momentum
4.73
Adaptive
4.64
Agency
4.90
Time
4.41
Anti-Purple
4.42
Anti-Repeat
4.25
Show > Tell
4.63
Subtext
4.63
Pacing
4.56
Quality trajectory ↑ builds
Early
4.21/5
Mid
4.52/5
Late
4.71/5
Behavioral · computed from this model's prose
Avg words / reply
312
+8 vs pool
Unique-word ratio
0.614
-0.007 vs pool
Bigram repetition
0.049
-0.015 vs pool
Flaw Hunter
Score (mean)
69.5
median 71.0
Fatal / session
0.07
Major / session
2.77
Sessions
56
Top flaws: purple prose · recycled description · narrating emotions
From the Round 03 flaw-hunter catch-up run. Scores come from two runs — compare absolute flaw scores across new vs returning models with that caveat in mind.
This is the judges' call — not the crowd's.
MiMo 2.5 Pro's scores here are judge-only. Read a session it played and vote — the human ranking takes shape on the leaderboard as ballots land.
Source: rp-benchmark · scripts/build-round3-judge-preview.py. Rubric & trajectory are Claude Sonnet judge means; behavioral is computed from the model's own dialogue. No human votes included.
← Back to the Round 03 standings