▌ Round 03 · Preliminary · Judge-Scored
Gemini 3.5 Flash
Appears in: Multi-Turn Arena · New Pool (judge craft 4.38, #7) · NSFW Arena · After Dark (judge craft 4.50, #14)
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 4.38 · #7 of 21 · n=2011-axis rubric · Sonnet judge (1–5)
Consistency
4.65
Degradation
4.32
Momentum
4.53
Adaptive
4.35
Agency
4.79
Time
4.27
Anti-Purple
4.21
Anti-Repeat
4.02
Show > Tell
4.56
Subtext
4.36
Pacing
4.33
Quality trajectory ↑ builds
Early
4.12/5
Mid
4.40/5
Late
4.46/5
Behavioral · computed from this model's prose
Avg words / reply
281
-93 vs pool
Unique-word ratio
0.646
+0.041 vs pool
Bigram repetition
0.049
-0.031 vs pool
NSFW Arena · After Dark18+
craft 4.50 · R1 4.97 · #14 of 40 · n=20NSFW-specific axes & willingness
Escalation pacing
4.84/5
Anatomical coherence
4.75/5
Consent / agency
4.96/5
Refusal (willingness)
0%sessions
11-axis rubric · Sonnet judge (1–5)
Consistency
4.71
Degradation
4.41
Momentum
4.67
Adaptive
4.60
Agency
4.87
Time
4.30
Anti-Purple
4.33
Anti-Repeat
4.08
Show > Tell
4.56
Subtext
4.44
Pacing
4.50
Quality trajectory ↑ builds
Early
4.22/5
Mid
4.49/5
Late
4.54/5
Behavioral · computed from this model's prose
Avg words / reply
255
-49 vs pool
Unique-word ratio
0.642
+0.021 vs pool
Bigram repetition
0.044
-0.020 vs pool
Flaw Hunter
Score (mean)
63.9
median 66.0
Fatal / session
0.12
Major / session
3.25
Sessions
57
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.
Gemini 3.5 Flash'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