▌ Round 03 · Preliminary · Judge-Scored
DeepSeek V4 Pro
Appears in: NSFW Arena · After Dark (judge craft 4.62, #4)
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.
NSFW Arena · After Dark18+
craft 4.62 · R1 4.98 · #4 of 40 · n=20NSFW-specific axes & willingness
Escalation pacing
4.83/5
Anatomical coherence
4.81/5
Consent / agency
4.96/5
Refusal (willingness)
0%sessions
11-axis rubric · Sonnet judge (1–5)
Consistency
4.79
Degradation
4.65
Momentum
4.74
Adaptive
4.64
Agency
4.89
Time
4.44
Anti-Purple
4.49
Anti-Repeat
4.33
Show > Tell
4.66
Subtext
4.65
Pacing
4.60
Quality trajectory ↑ builds
Early
4.23/5
Mid
4.53/5
Late
4.72/5
Behavioral · computed from this model's prose
Avg words / reply
253
-51 vs pool
Unique-word ratio
0.650
+0.029 vs pool
Bigram repetition
0.038
-0.026 vs pool
Flaw Hunter
Score (mean)
19.4
median 46.5
Fatal / session
0.50
Major / session
9.00
Sessions
8
Top flaws: purple prose · recycled description · narrating emotions
From the Round 01/02 flaw-hunter pool. 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.
DeepSeek V4 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