Benchmarks › Lech Mazur Short-Story

Lech Mazur Short-Story

Write short stories under the official writing constraints
3.8top score · Claude Fable 5.1
44models ranked
15open-weights models
Thurstone comparison scorehigher is better
Oct 20, 2026last updated

How the test works

1

Prompts

Write short stories under the official writing constraints

  • Kinds of tasksConstrained short-story writing
2

Votes

Blinded LLM pairwise judgments, both story orders

3

Rating

Thurstone comparison score

  • Better isHigher
Sources: github.com

Lech Mazur Short-Story leaderboardtop 10 of 44

Full leaderboard with filters →
ModelLech Mazur Short-StoryPer taskPriceSpeedSupportsContext
3.8high
$3.91$10 / $50581M
Claude Opus 5Anthropic
3.7xhigh
$4.88$5 / $25531M
3.7high
$1.82$4 / $20821M
3.4high
$1.73$10 / $50591.05M
GLM 5.3Z.aiOpen
3max
$2.01$1.40 / $4.401861M
2.9high
—$10 / $50—1M
2.5xhigh
$1.18$4 / $20851.05M
GPT-5.5OpenAI
2.4xhigh
$2.63$5 / $301131.05M
Kimi K3Moonshot AIOpen
2.4unstated
—$3 / $15—1M
GPT-5.4OpenAI
2medium
—$2.50 / $15—1.05M

Lech Mazur Short-Story questions

What is Lech Mazur Short-Story?›

Write short stories under the official writing constraints. It has 400 fixed writing briefs. Created by Independent project (Lech Mazur).

Which model is best on Lech Mazur Short-Story?›

Claude Fable 5.1 leads Lech Mazur Short-Story with 3.8, ahead of Claude Opus 5 (3.7) and Claude Opus 5.5 (3.7).

What is the best open-weights model on Lech Mazur Short-Story?›

GLM 5.3 with 3, then Kimi K3 (2.4).

How is Lech Mazur Short-Story scored?›

Thurstone comparison score; blinded llm pairwise judgments, both story orders; relative ratings correct presentation order.