llmboard.ai
Benchmarks
CompareRankings
llmboard.ai
Benchmarks
CompareRankings
HomeBenchmarkscreativityCreative Writing v3

creativity benchmark

Creative Writing v3

EQ-Bench Creative Writing v3 is an LLM-judged creative writing benchmark that evaluates models across 32 writing prompts with 3 iterations per prompt. Uses a hybrid scoring system combining rubric assessment and Elo ratings through pairwise comparisons. Challenges models in areas like humor, romance, spatial awareness, and unique perspectives to assess emotional intelligence and creative writing capabilities.

Updated Aug 7, 2026

Published models13
Registry coverage13
MetricScore
EvidenceB

On this page

  • Ranking
  • Distribution
  • Highlights
  • About
  • FAQ

Creative Writing v3 leaderboard

Sorted by the source-provided rank. Higher score is better according to the registry.

13 rows
Columns

Show columns

1ACQwen3-235B-A22B-Instruct-2507Alibaba Cloud / Qwen Team87.5%100.0%13CAug 7, 2026
2ACQwen3 VL 235B A22B InstructAlibaba Cloud / Qwen Team86.5%91.7%13CAug 7, 2026
3ACQwen3-235B-A22B-Thinking-2507Alibaba Cloud / Qwen Team86.1%83.3%13CAug 7, 2026
4XAGrok-4.1 ThinkingxAI86.1%75.0%13CAug 7, 2026
5ACQwen3 VL 235B A22B ThinkingAlibaba Cloud / Qwen Team85.7%66.7%13CAug 7, 2026
6ACQwen3 VL 32B InstructAlibaba Cloud / Qwen Team85.6%58.3%13CAug 7, 2026
7XAGrok-4.1xAI85.4%50.0%13CAug 7, 2026
8ACQwen3-Next-80B-A3B-InstructAlibaba Cloud / Qwen Team85.3%41.7%13CAug 7, 2026
9ACQwen3 VL 30B A3B InstructAlibaba Cloud / Qwen Team84.6%33.3%13CAug 7, 2026
10ACQwen3 VL 32B ThinkingAlibaba Cloud / Qwen Team83.3%25.0%13CAug 7, 2026
11ACQwen3 VL 30B A3B ThinkingAlibaba Cloud / Qwen Team82.5%16.7%13CAug 7, 2026
12ACQwen3 VL 8B ThinkingAlibaba Cloud / Qwen Team82.4%8.3%13CAug 7, 2026
13ACQwen3 VL 4B ThinkingAlibaba Cloud / Qwen Team76.1%0.0%13CAug 7, 2026

Score distribution

Top published rows on the benchmark's original scale.

Creative Writing v3

Creative Writing v3 highlights

The top published results on this benchmark's own scale.

Rank #1Qwen3-235B-A22B-Instruct-250787.5%Rank #2Qwen3 VL 235B A22B Instruct86.5%Rank #3Qwen3-235B-A22B-Thinking-250786.1%Rank #4Grok-4.1 Thinking86.1%

What is Creative Writing v3?

Definition and scoring fields from the benchmark registry.

EQ-Bench Creative Writing v3 is an LLM-judged creative writing benchmark that evaluates models across 32 writing prompts with 3 iterations per prompt. Uses a hybrid scoring system combining rubric assessment and Elo ratings through pairwise comparisons. Challenges models in areas like humor, romance, spatial awareness, and unique perspectives to assess emotional intelligence and creative writing capabilities.

Scores are shown in ratio. The current registry marks this benchmark as not independently verified with evidence level B.

Family
Creative Writing v3
Modality
text
Primary category
creativity
Score direction
higher
LLMBoard eligible
Yes
Evaluation key
creative-writing-v3|llm-stats-current

Source-native results are preserved. Eligibility for the overall LLMBoard score is a separate policy decision.

FAQ

Common questions about Creative Writing v3.

Which model scores highest on Creative Writing v3?

Qwen3-235B-A22B-Instruct-2507 is currently ranked first with 87.5%.

What does Creative Writing v3 measure?

EQ-Bench Creative Writing v3 is an LLM-judged creative writing benchmark that evaluates models across 32 writing prompts with 3 iterations per prompt. Uses a hybrid scoring system combining rubric assessment and Elo ratings through pairwise comparisons. Challenges models in areas like humor, romance, spatial awareness, and unique perspectives to assess emotional intelligence and creative writing capabilities.

Is a higher score better?

Yes. Higher values rank better for this benchmark.

How many models are compared?

13 unique published model results are currently shown.

Does this benchmark affect the overall score?

This benchmark is marked as eligible for the current LLMBoard capability methodology.

Rankings

OverallCodingText ArenaPricing

Modalities

Image GenerationVideo GenerationSpeech-to-TextEmbeddings

Benchmarks

All BenchmarksReasoningMathCoding

Vendors

All VendorsOpenAIAnthropicGoogle
llmboard.aiCopyright 2026 llmboard.ai