BasedAGIBasedAGI
Menu
Rankings live

creative

NPC dialogue

Low-latency in-character dialogue suitable for games.

#1 Recommendation

qwen-2.5-72b-instruct

Strong on Creative Writing Official (EQ-Bench Slice) creative_writing_score (78%) and Judgemark Official (EQ-Bench Slice) judgemark_score (56%)

external/qwen/qwen-2-5-72b-instruct

21.7%

Score

35.0%

Confidence

Limited benchmark evidence for this use case.

41 ranked models with average evidence of 11.5 points. Rankings may shift as more benchmark data is ingested.

Ranked Models

30

Evidence Quality

79%

Scoring

Benchmark-backed

Top Signal

Creative Writing Official (EQ-Bench Slice): creative_writing_score

All Ranked Models

Max params:
Min confidence:
30 of 30
RankModelScore
#12qwen-2.5-72b-instruct
21.7%
#16gpt-4o
20.7%
#24gemini-2.5-pro
18.5%
#35Grok-4-0709
17.1%
#40gpt-4.1-20250414
16.6%
#43Arch-Agent-32B
16.4%
#59gemma-2-27b-it
13.5%
#75xai-org/grok-4-fast-reasoning
12.1%
#82Arch-Agent-3B
11.6%
#86xai-org/grok-4-1-fast-reasoning
11.4%
#87gemini-3-pro-preview
11.4%
#90Arch-Agent-1.5B
11.1%
#95gemini-3-flash-preview
10.7%
#97x-ai/grok-3
10.6%
#99google/gemini-3.1-pro-preview
10.4%
#100claude-sonnet-4-20250514
10.4%
#103gemini-2.5-flash
10.2%
#113gpt-5-2025-08-07
9.5%
#115openai/gpt-5.4-2026-03-05
9.4%
#118gpt-5.1-2025-11-13
9.1%
#123anthropic/claude-sonnet-4.6
9.0%
#125claude-opus-4-5-20251101
9.0%
#129gpt-5-mini-2025-08-07
8.8%
#131xai-org/grok-4-1-fast-non-reasoning
8.7%
#132Kimi-K2-Instruct
8.7%
#135gpt-4o-2024-05-13
8.4%
#142xai-org/grok-4-fast-non-reasoning
8.2%
#147qwen/qwen3-max
7.9%
#153DeepSeek-V2.5
7.8%
#154Meta-Llama-3-70B-Instruct
7.8%

Compare Models

Model A leads by +1.0%

Shareable Link →

Model A

qwen-2.5-72b-instruct

external/qwen/qwen-2-5-72b-instruct

21.7%

Rank #12

Confidence 35.0%13 evidence pts

Creative Writing Official (EQ-Bench Slice): creative_writing_score

Value 78.4% · Conf 100.0% · Weight 6.1%

artificialanalysis_creative_writing_official.creative_writing_score (Mar 12, 2026)

Judgemark Official (EQ-Bench Slice): judgemark_score

Value 55.6% · Conf 100.0% · Weight 3.3%

artificialanalysis_judgemark_official.judgemark_score (Mar 12, 2026)

EQ-Bench Leaderboard: judgemark_score

Value 55.6% · Conf 100.0% · Weight 1.5%

eq_bench.judgemark_score (Mar 12, 2026)

Galileo Agent Leaderboard v2: Avg AC

Value 76.1% · Conf 100.0% · Weight 1.5%

galileo_agent_v2.avg_ac (Mar 12, 2026)

Model B

gpt-4o

external/openai/gpt-4o

20.7%

Rank #16

Confidence 26.5%12 evidence pts

Creative Writing Official (EQ-Bench Slice): creative_writing_score

Value 84.4% · Conf 100.0% · Weight 6.6%

artificialanalysis_creative_writing_official.creative_writing_score (Mar 12, 2026)

Judgemark Official (EQ-Bench Slice): judgemark_score

Value 74.3% · Conf 100.0% · Weight 4.4%

artificialanalysis_judgemark_official.judgemark_score (Mar 12, 2026)

EQ-Bench Leaderboard: judgemark_score

Value 74.3% · Conf 100.0% · Weight 2.0%

eq_bench.judgemark_score (Mar 12, 2026)

MEGA-Bench: overall_score

Value 92.8% · Conf 100.0% · Weight 0.7%

mega_bench.overall_score (Mar 12, 2026)

Ranking Diagnostics & Missing Models

Source Lift

Ranked

41

Sources

8

Quality

Insufficient

Vals CorpFin v2

vals_corp_fin_v2

22 rows

0.5% avg lift

Vals Legal Bench

vals_legal_bench

21 rows

0.5% avg lift

Vals GPQA

vals_gpqa

21 rows

0.4% avg lift

Vals Tax Eval v2

vals_tax_eval_v2

20 rows

0.5% avg lift

Missing Strong Models

gpt-5.2-2025-12-11

external/openai/gpt-5-2-2025-12-11

Rank #16

16.2%

Thin evidence after weighting

anthropic/claude-opus-4-6-thinking

external/anthropic/claude-opus-4-6-thinking

Rank #17

16.1%

Thin evidence after weighting

anthropic/claude-opus-4-5-20251101-thinking

external/anthropic/claude-opus-4-5-20251101-thinking

Rank #21

15.2%

Thin evidence after weighting

anthropic/claude-sonnet-4-5-20250929-thinking

external/anthropic/claude-sonnet-4-5-20250929-thinking

Rank #28

14.1%

Thin evidence after weighting
Taxonomy Details

Core Tasks

task.dialogue_character_voicetask.persona_consistency

Required Modes

mode.low_latencymode.persona_memory

Domains

domain.gaming_npc_dialogue

Related Use Cases