Fiction Bench (Story Hacker)
Models write three chapters for each of 20 genres. The overall score weights blind reader preferences at 60 percent and model judges’ craft scores at 40 percent.
The organizer’s leaderboard has scores for 42 models that completed all 20 stories. Human ratings with fewer than 20 votes are provisional.
- Made by
- Story Hacker AI
- Released
- October 2026
- Kind
- Creative writing
- Level
- University
- Scored
- Ranked by votes
- Answers
- Open-ended
- Tasks
- All public
- Measures
- Preference
- Subject
- Languages and literature
- Language
- English
- Interaction
- Single turn
- Items from
- Live play or votes
- Refresh
- One fixed set
- Run by
- A host
- Strongest model tested
- GPT-6.2 Astra #1 on the capability index
- Last model added
- November 2026
- Scores refreshed
- November 20th, 2026
GPT-6.2 Astra leads with 73.6 GPT-6.2 Astra leads with 73.6
Best score per entry from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran with several settings or harnesses, we keep its best. Every run is in the next chart. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. This view shows only models we track, each at its best.
- 73.6 GPT-6.2 Astra
- 72.6 GPT-6.2 Sol
- 68.3 GPT-6.1 Terra
- 66.3 Claude Opus 5.5
- 65.1
Gemini 4 Argon
- 64.4
Gemini 4 Carbon
- 64.0
Muse Wonder
- 63.6 Claude Fable 5.1
- 63.0 Claude Sonnet 5.5
- 62.7 GPT-6 Astra
- 62.5 Kimi K3.1 Pro
- 62.3 GPT-6 Sol
- 62.2 GPT-6 Astra Pro
- 62.1 Claude Fable 5.2
- 61.3 GPT-6 Luna
- 61.0
Muse Spark 1.3
- 60.7 Claude Opus 4.8
- 60.3
Gemini 3.7 Flash
- 60.0 GPT-6.1 Sol
- 59.7
Gemini 3.8 Flash
- 59.6 aion-labs/aion-3.5
- 58.5 Kimi K3
- 57.3 aion-labs/aion-3.0
- 56.9 aion-labs/aion-3.5-mini
- 56.8
DeepSeek V4.1 Flash 1017
- 55.9
Qwen 3.8 Max
- 55.2 GLM-5.3-Flash
- 55.2 MiMo-V2.6 Flash
- 54.9
Muse Ember
- 54.8 Claude Haiku 5.5
- 54.5
MiniMax-M3
- 54.1 Grok 4.7
- 53.9 MiMo-V2.6 Pro
- 52.3 GLM-5.3
- 52.2
Gemini 3.5 Flash-Lite
- 51.6
DeepSeek V4 Pro 0813
- 51.6
DeepSeek V4.1 Flash
- 51.2 aion-labs/aion-3.0-mini
- 50.0
Muse Glimmer
- 49.6 aion-labs/aion-2.0
- 46.1
Qwen3.8 Flash
- 43.2 mistralai/mistral-large-4-0
- 1 GPT-6.2 Astra Provisional human rating 73.6
- 2 GPT-6.2 Sol Provisional human rating 72.6
- 3 GPT-6.1 Terra 68.3
- 4 Claude Opus 5.5 66.3
- 5
Gemini 4 Argon 65.1
- 6
Gemini 4 Carbon 64.4
- 7
Muse Wonder 64.0
- 8 Claude Fable 5.1 63.6
- 9 Claude Sonnet 5.5 63.0
- 10 GPT-6 Astra 62.7
- 11 Kimi K3.1 Pro 62.5
- 12 GPT-6 Sol 62.3
- 13 GPT-6 Astra Pro 62.2
- 14 Claude Fable 5.2 62.1
- 15 GPT-6 Luna 61.3
- 16
Muse Spark 1.3 61.0
- 17 Claude Opus 4.8 60.7
- 18
Gemini 3.7 Flash 60.3
- 19 GPT-6.1 Sol 60.0
- 20
Gemini 3.8 Flash 59.7
- 21 aion-labs/aion-3.5 59.6
- 22 Kimi K3 58.5
- 23 aion-labs/aion-3.0 57.3
- 24 aion-labs/aion-3.5-mini 56.9
- 25
DeepSeek V4.1 Flash 1017 56.8
- 26
Qwen 3.8 Max prime 55.9
- 27 GLM-5.3-Flash 55.2
- 28 MiMo-V2.6 Flash 55.2
- 29
Muse Ember 54.9
- 30 Claude Haiku 5.5 54.8
- 31
MiniMax-M3 54.5
- 32 Grok 4.7 54.1
- 33 MiMo-V2.6 Pro 53.9
- 34 GLM-5.3 52.3
- 35
Gemini 3.5 Flash-Lite 52.2
- 36
DeepSeek V4 Pro 0813 51.6
- 37
DeepSeek V4.1 Flash 51.6
- 38 aion-labs/aion-3.0-mini 51.2
- 39
Muse Glimmer 50.0
- 40 aion-labs/aion-2.0 49.6
- 41
Qwen3.8 Flash 46.1
- 42 mistralai/mistral-large-4-0 43.2
Source: Official Fiction Bench (Story Hacker) leaderboard
- 73.6 GPT-6.2 Astra
- 72.6 GPT-6.2 Sol
- 68.3 GPT-6.1 Terra
- 66.3 Claude Opus 5.5
- 65.1
Gemini 4 Argon
- 64.4
Gemini 4 Carbon
- 64.0
Muse Wonder
- 63.6 Claude Fable 5.1
- 63.0 Claude Sonnet 5.5
- 62.7 GPT-6 Astra
- 62.5 Kimi K3.1 Pro
- 62.3 GPT-6 Sol
- 62.2 GPT-6 Astra Pro
- 62.1 Claude Fable 5.2
- 61.3 GPT-6 Luna
- 61.0
Muse Spark 1.3
- 60.7 Claude Opus 4.8
- 60.3
Gemini 3.7 Flash
- 60.0 GPT-6.1 Sol
- 59.7
Gemini 3.8 Flash
- 58.5 Kimi K3
- 56.8
DeepSeek V4.1 Flash 1017
- 55.9
Qwen 3.8 Max
- 55.2 GLM-5.3-Flash
- 55.2 MiMo-V2.6 Flash
- 54.9
Muse Ember
- 54.8 Claude Haiku 5.5
- 54.5
MiniMax-M3
- 54.1 Grok 4.7
- 53.9 MiMo-V2.6 Pro
- 52.3 GLM-5.3
- 52.2
Gemini 3.5 Flash-Lite
- 51.6
DeepSeek V4 Pro 0813
- 51.6
DeepSeek V4.1 Flash
- 50.0
Muse Glimmer
- 46.1
Qwen3.8 Flash
- 1 GPT-6.2 Astra Provisional human rating 73.6
- 2 GPT-6.2 Sol Provisional human rating 72.6
- 3 GPT-6.1 Terra 68.3
- 4 Claude Opus 5.5 66.3
- 5
Gemini 4 Argon 65.1
- 6
Gemini 4 Carbon 64.4
- 7
Muse Wonder 64.0
- 8 Claude Fable 5.1 63.6
- 9 Claude Sonnet 5.5 63.0
- 10 GPT-6 Astra 62.7
- 11 Kimi K3.1 Pro 62.5
- 12 GPT-6 Sol 62.3
- 13 GPT-6 Astra Pro 62.2
- 14 Claude Fable 5.2 62.1
- 15 GPT-6 Luna 61.3
- 16
Muse Spark 1.3 61.0
- 17 Claude Opus 4.8 60.7
- 18
Gemini 3.7 Flash 60.3
- 19 GPT-6.1 Sol 60.0
- 20
Gemini 3.8 Flash 59.7
- 21 Kimi K3 58.5
- 22
DeepSeek V4.1 Flash 1017 56.8
- 23
Qwen 3.8 Max prime 55.9
- 24 GLM-5.3-Flash 55.2
- 25 MiMo-V2.6 Flash 55.2
- 26
Muse Ember 54.9
- 27 Claude Haiku 5.5 54.8
- 28
MiniMax-M3 54.5
- 29 Grok 4.7 54.1
- 30 MiMo-V2.6 Pro 53.9
- 31 GLM-5.3 52.3
- 32
Gemini 3.5 Flash-Lite 52.2
- 33
DeepSeek V4 Pro 0813 51.6
- 34
DeepSeek V4.1 Flash 51.6
- 35
Muse Glimmer 50.0
- 36
Qwen3.8 Flash 46.1
Source: Official Fiction Bench (Story Hacker) leaderboard
| # | Entry | Score | Setting | Released | Run on | Source |
|---|---|---|---|---|---|---|
| 1 | GPT-6.2 Astra | 73.6 | Provisional human rating | Nov 2026 | Nov 2026 | Official leaderboard |
| 2 | GPT-6.2 Sol | 72.6 | Provisional human rating | Nov 2026 | Nov 2026 | Official leaderboard |
| 3 | GPT-6.1 Terra | 68.3 | Oct 2026 | Oct 2026 | Official leaderboard | |
| 4 | Claude Opus 5.5 | 66.3 | Sep 2026 | Oct 2026 | Official leaderboard | |
| 5 | | 65.1 | Sep 2026 | Oct 2026 | Official leaderboard | |
| 6 | | 64.4 | Oct 2026 | Nov 2026 | Official leaderboard | |
| 7 | | 64.0 | Oct 2026 | Oct 2026 | Official leaderboard | |
| 8 | Claude Fable 5.1 | 63.6 | Sep 2026 | Oct 2026 | Official leaderboard | |
| 9 | Claude Sonnet 5.5 | 63.0 | Sep 2026 | Oct 2026 | Official leaderboard | |
| 10 | GPT-6 Astra | 62.7 | Sep 2026 | Oct 2026 | Official leaderboard | |
| 11 | Kimi K3.1 Pro | 62.5 | Oct 2026 | Oct 2026 | Official leaderboard | |
| 12 | GPT-6 Sol | 62.3 | Sep 2026 | Oct 2026 | Official leaderboard | |
| 13 | GPT-6 Astra Pro | 62.2 | Sep 2026 | Oct 2026 | Official leaderboard | |
| 14 | Claude Fable 5.2 | 62.1 | Nov 2026 | Nov 2026 | Official leaderboard | |
| 15 | GPT-6 Luna | 61.3 | Sep 2026 | Oct 2026 | Official leaderboard | |
| 16 | | 61.0 | Sep 2026 | Oct 2026 | Official leaderboard | |
| 17 | Claude Opus 4.8 | 60.7 | May 2026 | Oct 2026 | Official leaderboard | |
| 18 | | 60.3 | Aug 2026 | Oct 2026 | Official leaderboard | |
| 19 | GPT-6.1 Sol | 60.0 | Sep 2026 | Oct 2026 | Official leaderboard | |
| 20 | | 59.7 | Sep 2026 | Oct 2026 | Official leaderboard | |
| 21 | aion-labs/aion-3.5 | 59.6 | Oct 2026 | Official leaderboard | ||
| 22 | Kimi K3 | 58.5 | Jul 2026 | Oct 2026 | Official leaderboard | |
| 23 | aion-labs/aion-3.0 | 57.3 | Oct 2026 | Official leaderboard | ||
| 24 | aion-labs/aion-3.5-mini | 56.9 | Oct 2026 | Official leaderboard | ||
| 25 | | 56.8 | Oct 2026 | Oct 2026 | Official leaderboard | |
| 26 | | 55.9 | prime | Aug 2026 | Oct 2026 | Official leaderboard |
| 27 | GLM-5.3-Flash | 55.2 | Aug 2026 | Oct 2026 | Official leaderboard | |
| 28 | MiMo-V2.6 Flash | 55.2 | Sep 2026 | Oct 2026 | Official leaderboard | |
| 29 | | 54.9 | Nov 2026 | Nov 2026 | Official leaderboard | |
| 30 | Claude Haiku 5.5 | 54.8 | Oct 2026 | Oct 2026 | Official leaderboard | |
| 31 | | 54.5 | Jun 2026 | Oct 2026 | Official leaderboard | |
| 32 | Grok 4.7 | 54.1 | Sep 2026 | Oct 2026 | Official leaderboard | |
| 33 | MiMo-V2.6 Pro | 53.9 | Sep 2026 | Oct 2026 | Official leaderboard | |
| 34 | GLM-5.3 | 52.3 | Aug 2026 | Oct 2026 | Official leaderboard | |
| 35 | | 52.2 | Jul 2026 | Oct 2026 | Official leaderboard | |
| 36 | | 51.6 | Aug 2026 | Oct 2026 | Official leaderboard | |
| 37 | | 51.6 | Sep 2026 | Oct 2026 | Official leaderboard | |
| 38 | aion-labs/aion-3.0-mini | 51.2 | Oct 2026 | Official leaderboard | ||
| 39 | | 50.0 | Aug 2026 | Oct 2026 | Official leaderboard | |
| 40 | aion-labs/aion-2.0 | 49.6 | Oct 2026 | Official leaderboard | ||
| 41 | | 46.1 | Aug 2026 | Oct 2026 | Official leaderboard | |
| 42 | mistralai/mistral-large-4-0 | 43.2 | Oct 2026 | Official leaderboard |
| # | Model | Score | Setting | Released | Run on | Source |
|---|---|---|---|---|---|---|
| 1 | GPT-6.2 Astra | 73.6 | Provisional human rating | Nov 2026 | Nov 2026 | Official leaderboard |
| 2 | GPT-6.2 Sol | 72.6 | Provisional human rating | Nov 2026 | Nov 2026 | Official leaderboard |
| 3 | GPT-6.1 Terra | 68.3 | Oct 2026 | Oct 2026 | Official leaderboard | |
| 4 | Claude Opus 5.5 | 66.3 | Sep 2026 | Oct 2026 | Official leaderboard | |
| 5 | | 65.1 | Sep 2026 | Oct 2026 | Official leaderboard | |
| 6 | | 64.4 | Oct 2026 | Nov 2026 | Official leaderboard | |
| 7 | | 64.0 | Oct 2026 | Oct 2026 | Official leaderboard | |
| 8 | Claude Fable 5.1 | 63.6 | Sep 2026 | Oct 2026 | Official leaderboard | |
| 9 | Claude Sonnet 5.5 | 63.0 | Sep 2026 | Oct 2026 | Official leaderboard | |
| 10 | GPT-6 Astra | 62.7 | Sep 2026 | Oct 2026 | Official leaderboard | |
| 11 | Kimi K3.1 Pro | 62.5 | Oct 2026 | Oct 2026 | Official leaderboard | |
| 12 | GPT-6 Sol | 62.3 | Sep 2026 | Oct 2026 | Official leaderboard | |
| 13 | GPT-6 Astra Pro | 62.2 | Sep 2026 | Oct 2026 | Official leaderboard | |
| 14 | Claude Fable 5.2 | 62.1 | Nov 2026 | Nov 2026 | Official leaderboard | |
| 15 | GPT-6 Luna | 61.3 | Sep 2026 | Oct 2026 | Official leaderboard | |
| 16 | | 61.0 | Sep 2026 | Oct 2026 | Official leaderboard | |
| 17 | Claude Opus 4.8 | 60.7 | May 2026 | Oct 2026 | Official leaderboard | |
| 18 | | 60.3 | Aug 2026 | Oct 2026 | Official leaderboard | |
| 19 | GPT-6.1 Sol | 60.0 | Sep 2026 | Oct 2026 | Official leaderboard | |
| 20 | | 59.7 | Sep 2026 | Oct 2026 | Official leaderboard | |
| 21 | Kimi K3 | 58.5 | Jul 2026 | Oct 2026 | Official leaderboard | |
| 22 | | 56.8 | Oct 2026 | Oct 2026 | Official leaderboard | |
| 23 | | 55.9 | prime | Aug 2026 | Oct 2026 | Official leaderboard |
| 24 | GLM-5.3-Flash | 55.2 | Aug 2026 | Oct 2026 | Official leaderboard | |
| 25 | MiMo-V2.6 Flash | 55.2 | Sep 2026 | Oct 2026 | Official leaderboard | |
| 26 | | 54.9 | Nov 2026 | Nov 2026 | Official leaderboard | |
| 27 | Claude Haiku 5.5 | 54.8 | Oct 2026 | Oct 2026 | Official leaderboard | |
| 28 | | 54.5 | Jun 2026 | Oct 2026 | Official leaderboard | |
| 29 | Grok 4.7 | 54.1 | Sep 2026 | Oct 2026 | Official leaderboard | |
| 30 | MiMo-V2.6 Pro | 53.9 | Sep 2026 | Oct 2026 | Official leaderboard | |
| 31 | GLM-5.3 | 52.3 | Aug 2026 | Oct 2026 | Official leaderboard | |
| 32 | | 52.2 | Jul 2026 | Oct 2026 | Official leaderboard | |
| 33 | | 51.6 | Aug 2026 | Oct 2026 | Official leaderboard | |
| 34 | | 51.6 | Sep 2026 | Oct 2026 | Official leaderboard | |
| 35 | | 50.0 | Aug 2026 | Oct 2026 | Official leaderboard | |
| 36 | | 46.1 | Aug 2026 | Oct 2026 | Official leaderboard |
Rankings by part
We read 25 parts from the benchmark’s own leaderboard, each ranking models on one part of the test, such as Story, Intelligence and Prose quality. Claude Fable 5.2 leads 18 of them, and 2 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.
Story
36 models we track have a score here.
- 1 Claude Fable 5.2 ↑13 84.1
- 2 GPT-6.2 Astra ↓1 81.1
- 3 Claude Opus 5.5 ↑1 81.1
- 4 GPT-6.2 Sol ↓2 79.1
- 5 Claude Fable 5.1 ↑3 79.0
- 6 GPT-6 Astra Pro ↑7 78.2
- 7 Claude Sonnet 5.5 ↑2 78.2
- 8 GPT-6 Astra ↑2 78.1
- 9
Muse Wonder ↓2 77.2
- 10 Kimi K3.1 Pro ↑1 77.1
Intelligence
36 models we track have a score here.
- 1 Claude Fable 5.2 ↑13 89.3
- 2 Claude Opus 5.5 ↑2 86.2
- 3 GPT-6.2 Astra ↓2 85.3
- 4 GPT-6 Astra Pro ↑9 84.8
- 5 GPT-6 Astra ↑5 84.4
- 6 GPT-6.2 Sol ↓4 84.2
- 7 Claude Fable 5.1 ↑1 83.9
- 8 Kimi K3.1 Pro ↑3 83.7
- 9 Claude Sonnet 5.5 83.0
- 10 GPT-6.1 Sol ↑9 83.0
Prose quality
36 models we track have a score here.
- 1 GPT-6.2 Astra 90.9
- 2 Claude Fable 5.2 ↑12 90.5
- 3 GPT-6.2 Sol ↓1 90.5
- 4 GPT-6 Astra ↑6 87.4
- 5 GPT-6 Astra Pro ↑8 87.4
- 6 Claude Opus 5.5 ↓2 86.2
- 7 GPT-6.1 Sol ↑12 85.6
- 8 Kimi K3.1 Pro ↑3 85.2
- 9 Claude Fable 5.1 ↓1 85.1
- 10 GPT-6 Sol ↑2 84.9
Human rating
36 models we track have a score here.
- 1 GPT-6.2 Astra 65.5
- 2 GPT-6.2 Sol 64.7
- 3 GPT-6.1 Terra 61.8
- 4
Gemini 4 Argon ↑1 58.5
- 5 GPT-6 Luna ↑10 56.5
- 6
Gemini 3.7 Flash ↑12 56.5
- 7
Gemini 4 Carbon ↓1 54.3
- 8 Claude Opus 5.5 ↓4 54.0
- 9
Muse Wonder ↓2 53.2
- 10
Gemini 3.8 Flash ↑10 52.6
AI score
36 models we track have a score here.
- 1 Claude Fable 5.2 ↑13 87.6
- 2 GPT-6.2 Astra ↓1 85.0
- 3 Claude Opus 5.5 ↑1 84.1
- 4 GPT-6.2 Sol ↓2 83.9
- 5 GPT-6 Astra Pro ↑8 83.2
- 6 GPT-6 Astra ↑4 83.0
- 7 Claude Fable 5.1 ↑1 81.9
- 8 Kimi K3.1 Pro ↑3 81.5
- 9 Claude Sonnet 5.5 81.2
- 10 GPT-6.1 Sol ↑9 81.0
Believable Character Actions
36 models we track have a score here.
- 1 Claude Fable 5.2 ↑13 85.2
- 2 GPT-6.2 Astra ↓1 85.0
- 3 GPT-6 Astra ↑7 82.5
- 4 Claude Opus 5.5 81.9
- 5 GPT-6 Astra Pro ↑8 81.8
- 6 GPT-6.2 Sol ↓4 81.7
- 7 Kimi K3.1 Pro ↑4 80.2
- 8 Claude Fable 5.1 79.5
- 9 Claude Sonnet 5.5 78.9
- 10 GPT-6.1 Sol ↑9 78.2
Nuanced Characters
36 models we track have a score here.
- 1 Claude Fable 5.2 ↑13 85.9
- 2 GPT-6.2 Astra ↓1 83.1
- 3 Claude Opus 5.5 ↑1 82.4
- 4 Claude Fable 5.1 ↑4 82.3
- 5 GPT-6.2 Sol ↓3 81.9
- 6 Claude Sonnet 5.5 ↑3 81.4
- 7
Muse Wonder 80.9
- 8 GPT-6 Astra Pro ↑5 80.3
- 9 Kimi K3.1 Pro ↑2 80.1
- 10 GPT-6 Astra 79.8
Genre Craft
36 models we track have a score here.
- 1 Claude Fable 5.2 ↑13 86.6
- 2 Claude Opus 5.5 ↑2 83.0
- 3 GPT-6 Astra ↑7 82.0
- 4 GPT-6 Astra Pro ↑9 81.9
- 5 GPT-6.2 Astra ↓4 81.9
- 6 GPT-6.2 Sol ↓4 80.3
- 7
Gemini 4 Carbon ↓1 80.1
- 8 Kimi K3.1 Pro ↑3 79.3
- 9 GPT-6.1 Sol ↑10 79.3
- 10 Claude Fable 5.1 ↓2 78.3
Unsurprising or Uncreative
36 models we track have a score here.
- 1 GPT-6.2 Astra 81.5
- 2 Claude Fable 5.2 ↑12 81.4
- 3 GPT-6.2 Sol ↓1 79.1
- 4 Claude Fable 5.1 ↑4 78.3
- 5 Claude Sonnet 5.5 ↑4 77.9
- 6 Claude Opus 5.5 ↓2 77.4
- 7
Muse Wonder 74.8
- 8 GPT-6 Astra Pro ↑5 73.7
- 9 Kimi K3.1 Pro ↑2 73.4
- 10 GPT-6 Astra 73.4
World Building
36 models we track have a score here.
- 1 Claude Fable 5.2 ↑13 85.4
- 2 Claude Opus 5.5 ↑2 82.6
- 3
Gemini 4 Carbon ↑3 81.8
- 4 GPT-6.2 Astra ↓3 80.8
- 5 Claude Fable 5.1 ↑3 80.0
- 6 Claude Sonnet 5.5 ↑3 78.2
- 7
Muse Wonder 77.4
- 8 GPT-6.2 Sol ↓6 77.3
- 9 GPT-6 Astra ↑1 76.8
- 10
Gemini 4 Argon ↓5 76.3
Emotionally Engaging
36 models we track have a score here.
- 1 Claude Fable 5.2 ↑13 82.6
- 2 Claude Opus 5.5 ↑2 81.2
- 3 GPT-6.2 Astra ↓2 80.1
- 4 Claude Fable 5.1 ↑4 79.1
- 5 GPT-6.2 Sol ↓3 79.0
- 6
Muse Wonder ↑1 78.4
- 7 GPT-6 Astra Pro ↑6 77.9
- 8 Claude Sonnet 5.5 ↑1 77.8
- 9 GPT-6 Astra ↑1 77.1
- 10 Kimi K3.1 Pro ↑1 76.9
Compelling Plot
36 models we track have a score here.
- 1 Claude Fable 5.2 ↑13 81.3
- 2 Claude Opus 5.5 ↑2 78.8
- 3
Gemini 4 Carbon ↑3 76.6
- 4 Claude Fable 5.1 ↑4 75.8
- 5 GPT-6 Astra Pro ↑8 75.7
- 6 GPT-6 Astra ↑4 75.4
- 7 Claude Sonnet 5.5 ↑2 75.0
- 8 GPT-6.2 Astra ↓7 75.0
- 9
Muse Wonder ↓2 74.7
- 10 GPT-6.2 Sol ↓8 74.6
Followed Chapter Plan
36 models we track have a score here.
- 1 Claude Fable 5.2 ↑13 94.2
- 2 Claude Opus 5.5 ↑2 91.2
- 3 GPT-6.2 Sol ↓1 90.3
- 4 GPT-6.2 Astra ↓3 90.2
- 5 Kimi K3.1 Pro ↑6 89.4
- 6 Claude Sonnet 5.5 ↑3 89.0
- 7 Claude Fable 5.1 ↑1 88.8
- 8 GPT-6 Astra Pro ↑5 88.8
- 9
Gemini 4 Carbon ↓3 88.7
- 10 GPT-6 Astra 88.0
Unearned Transformations
36 models we track have a score here.
- 1 Claude Fable 5.2 ↑13 88.3
- 2 GPT-6.2 Astra ↓1 88.2
- 3 GPT-6 Astra Pro ↑10 86.2
- 4 Claude Opus 5.5 86.0
- 5 GPT-6 Astra ↑5 85.9
- 6 GPT-6.2 Sol ↓4 85.8
- 7 GPT-6.1 Sol ↑12 84.7
- 8 Claude Fable 5.1 84.4
- 9 Kimi K3.1 Pro ↑2 84.4
- 10 Claude Sonnet 5.5 ↓1 83.6
Faithful to Writing Prompt
36 models we track have a score here.
- 1 Claude Fable 5.2 ↑13 92.0
- 2 Claude Opus 5.5 ↑2 89.0
- 3 GPT-6 Astra Pro ↑10 89.0
- 4
Gemini 4 Carbon ↑2 88.8
- 5 GPT-6 Astra ↑5 88.3
- 6 Kimi K3.1 Pro ↑5 88.2
- 7 GPT-6.2 Sol ↓5 87.6
- 8 GPT-6.2 Astra ↓7 87.6
- 9 GPT-6.1 Sol ↑10 87.4
- 10 Claude Fable 5.1 ↓2 87.2
Characters Consistent with Profile
36 models we track have a score here.
- 1 Claude Fable 5.2 ↑13 91.9
- 2 Claude Opus 5.5 ↑2 88.4
- 3 Claude Fable 5.1 ↑5 87.8
- 4 GPT-6.2 Astra ↓3 87.2
- 5 Claude Sonnet 5.5 ↑4 87.1
- 6 Kimi K3.1 Pro ↑5 86.5
- 7 GPT-6.2 Sol ↓5 86.3
- 8
Gemini 4 Carbon ↓2 86.0
- 9 GPT-6 Astra Pro ↑4 85.8
- 10 GPT-6 Astra 85.8
Well-earned Lightness or Darkness
36 models we track have a score here.
- 1 Claude Fable 5.2 ↑13 87.1
- 2 Claude Opus 5.5 ↑2 84.7
- 3 GPT-6.2 Astra ↓2 84.4
- 4 GPT-6.2 Sol ↓2 83.0
- 5 GPT-6 Astra Pro ↑8 82.7
- 6 Claude Fable 5.1 ↑2 82.6
- 7 Claude Sonnet 5.5 ↑2 82.3
- 8 GPT-6 Astra ↑2 82.3
- 9 Kimi K3.1 Pro ↑2 81.9
- 10
Muse Wonder ↓3 81.4
Coherent
36 models we track have a score here.
- 1 Claude Fable 5.2 ↑13 87.4
- 2 GPT-6 Astra Pro ↑11 84.0
- 3 GPT-6 Astra ↑7 83.7
- 4 Claude Opus 5.5 82.9
- 5 GPT-6.1 Sol ↑14 81.4
- 6 GPT-6.2 Astra ↓5 80.8
- 7 Kimi K3.1 Pro ↑4 79.9
- 8 GPT-6.2 Sol ↓6 79.4
- 9 Claude Fable 5.1 ↓1 78.6
- 10 GPT-6 Sol ↑2 78.0
Pacing
36 models we track have a score here.
- 1 Claude Fable 5.2 ↑13 84.2
- 2 Claude Opus 5.5 ↑2 81.3
- 3 GPT-6.2 Astra ↓2 78.8
- 4 Claude Fable 5.1 ↑4 77.8
- 5 GPT-6.2 Sol ↓3 77.1
- 6 GPT-6 Astra Pro ↑7 77.0
- 7 GPT-6 Astra ↑3 76.8
- 8
Gemini 4 Carbon ↓2 76.8
- 9 Claude Sonnet 5.5 76.7
- 10 Kimi K3.1 Pro ↑1 75.8
Purple Prose
36 models we track have a score here.
- 1 GPT-6.2 Sol ↑1 95.6
- 2 GPT-6.2 Astra ↓1 95.3
- 3 Claude Fable 5.2 ↑11 94.6
- 4 GPT-6 Astra ↑6 94.4
- 5 GPT-6 Astra Pro ↑8 94.4
- 6 GPT-6.1 Terra ↓3 94.1
- 7 GPT-6.1 Sol ↑12 93.8
- 8 GPT-6 Sol ↑4 93.5
- 9 Kimi K3.1 Pro ↑2 92.5
- 10 Claude Opus 5.5 ↓6 91.5
Forced Poetry or Metaphor
36 models we track have a score here.
- 1 GPT-6.2 Sol ↑1 94.6
- 2 GPT-6.2 Astra ↓1 94.0
- 3 GPT-6.1 Terra 93.9
- 4 GPT-6 Astra ↑6 93.4
- 5 GPT-6 Astra Pro ↑8 93.3
- 6 GPT-6 Sol ↑6 93.0
- 7 GPT-6.1 Sol ↑12 92.8
- 8 Claude Fable 5.2 ↑6 92.4
- 9 Kimi K3.1 Pro ↑2 89.6
- 10 Claude Opus 5.5 ↓6 86.8
Slop
36 models we track have a score here.
- 1 GPT-6.2 Astra 93.5
- 2 GPT-6.2 Sol 91.6
- 3 Claude Fable 5.2 ↑11 91.2
- 4 GPT-6.1 Terra ↓1 85.6
- 5 GPT-6 Sol ↑7 85.4
- 6 Claude Opus 5.5 ↓2 85.2
- 7 GPT-6 Astra Pro ↑6 84.4
- 8 Claude Sonnet 5.5 ↑1 84.4
- 9 GPT-6 Astra ↑1 84.3
- 10 Claude Fable 5.1 ↓2 81.9
Show-Don't-Tell
36 models we track have a score here.
- 1 Claude Fable 5.2 ↑13 88.5
- 2 GPT-6.2 Sol 88.3
- 3 GPT-6.2 Astra ↓2 87.7
- 4 GPT-6 Astra ↑6 84.7
- 5 GPT-6 Astra Pro ↑8 84.4
- 6 Claude Opus 5.5 ↓2 84.3
- 7 Claude Fable 5.1 ↑1 84.0
- 8 Claude Sonnet 5.5 ↑1 82.6
- 9 GPT-6.1 Sol ↑10 82.4
- 10 Kimi K3.1 Pro ↑1 81.5
Amateurish
36 models we track have a score here.
- 1 Claude Fable 5.2 ↑13 89.2
- 2 GPT-6.2 Astra ↓1 87.5
- 3 GPT-6.2 Sol ↓1 86.2
- 4 Claude Opus 5.5 85.4
- 5 GPT-6 Astra Pro ↑8 85.1
- 6 GPT-6 Astra ↑4 84.9
- 7 Kimi K3.1 Pro ↑4 83.6
- 8 Claude Fable 5.1 83.2
- 9 GPT-6.1 Sol ↑10 82.4
- 10 Claude Sonnet 5.5 ↓1 82.1
Weak Dialogue
36 models we track have a score here.
- 1 GPT-6.2 Astra 87.3
- 2 Claude Fable 5.2 ↑12 87.1
- 3 GPT-6.2 Sol ↓1 86.5
- 4 Claude Fable 5.1 ↑4 84.3
- 5 Claude Opus 5.5 ↓1 83.9
- 6 Claude Sonnet 5.5 ↑3 83.6
- 7 GPT-6 Astra ↑3 82.6
- 8 GPT-6 Astra Pro ↑5 82.5
- 9 Kimi K3.1 Pro ↑2 82.3
- 10
Muse Wonder ↓3 80.9
The best score, 73.6, costs $8.78 for the full run
Runs on the cost frontier score higher than any run that's cheaper. GPT-6.2 Sol at Provisional human rating comes within 1 of the best for $2.37 for the full run, 4 times cheaper. The costs are the leaderboard's own estimates, so compare them only within this benchmark.
Source: Official Fiction Bench (Story Hacker) leaderboard
| Model | Effort | Harness | Score | Cost of the full run | Output tokens |
|---|---|---|---|---|---|
| GPT-6.2 Astra | Provisional human rating | 73.6 | $8.78 | ||
| GPT-6.2 Sol | Provisional human rating | 72.6 | $2.37 | ||
| GPT-6.1 Terra | 68.3 | $1.63 | |||
| Claude Opus 5.5 | 66.3 | $13 | |||
| 65.1 | $3.51 | ||||
| 64.4 | $9.19 | ||||
| 64.0 | $13 | ||||
| Claude Fable 5.1 | 63.6 | $32 | |||
| Claude Sonnet 5.5 | 63.0 | $7.43 | |||
| GPT-6 Astra | 62.7 | $14 | |||
| Kimi K3.1 Pro | 62.5 | $8.90 | |||
| GPT-6 Sol | 62.3 | $3.65 | |||
| GPT-6 Astra Pro | 62.2 | $43 | |||
| Claude Fable 5.2 | 62.1 | $42 | |||
| GPT-6 Luna | 61.3 | $0.21 | |||
| 61.0 | $2.68 | ||||
| Claude Opus 4.8 | 60.7 | $13 | |||
| 60.3 | $1.87 | ||||
| GPT-6.1 Sol | 60.0 | $2.74 | |||
| 59.7 | $2.88 | ||||
| Kimi K3 | 58.5 | $13 | |||
| 56.8 | $0.15 | ||||
| prime | 55.9 | $30 | |||
| GLM-5.3-Flash | 55.2 | $0.44 | |||
| MiMo-V2.6 Flash | 55.2 | $0.12 | |||
| 54.9 | $0.66 | ||||
| Claude Haiku 5.5 | 54.8 | $0.37 | |||
| 54.5 | $0.79 | ||||
| Grok 4.7 | 54.1 | $3.92 | |||
| MiMo-V2.6 Pro | 53.9 | $0.40 | |||
| GLM-5.3 | 52.3 | $8.77 | |||
| 52.2 | $0.68 | ||||
| 51.6 | $1.68 | ||||
| 51.6 | $0.43 | ||||
| 50.0 | $0.48 | ||||
| 46.1 | $0.88 |
The best score, 73.6, costs $8.78 for the full run
Runs on the cost frontier score higher than any run that's cheaper. GPT-6.2 Sol at Provisional human rating comes within 1 of the best for $2.37 for the full run, 4 times cheaper. The costs are the leaderboard's own estimates, so compare them only within this benchmark.
Source: Official Fiction Bench (Story Hacker) leaderboard
| Model | Effort | Harness | Score | Cost of the full run | Output tokens |
|---|---|---|---|---|---|
| GPT-6.2 Astra | Provisional human rating | 73.6 | $8.78 | ||
| GPT-6.2 Sol | Provisional human rating | 72.6 | $2.37 | ||
| GPT-6.1 Terra | 68.3 | $1.63 | |||
| Claude Opus 5.5 | 66.3 | $13 | |||
| 65.1 | $3.51 | ||||
| 64.4 | $9.19 | ||||
| 64.0 | $13 | ||||
| Claude Fable 5.1 | 63.6 | $32 | |||
| Claude Sonnet 5.5 | 63.0 | $7.43 | |||
| GPT-6 Astra | 62.7 | $14 | |||
| Kimi K3.1 Pro | 62.5 | $8.90 | |||
| GPT-6 Sol | 62.3 | $3.65 | |||
| GPT-6 Astra Pro | 62.2 | $43 | |||
| Claude Fable 5.2 | 62.1 | $42 | |||
| GPT-6 Luna | 61.3 | $0.21 | |||
| 61.0 | $2.68 | ||||
| Claude Opus 4.8 | 60.7 | $13 | |||
| 60.3 | $1.87 | ||||
| GPT-6.1 Sol | 60.0 | $2.74 | |||
| 59.7 | $2.88 | ||||
| Kimi K3 | 58.5 | $13 | |||
| 56.8 | $0.15 | ||||
| prime | 55.9 | $30 | |||
| GLM-5.3-Flash | 55.2 | $0.44 | |||
| MiMo-V2.6 Flash | 55.2 | $0.12 | |||
| 54.9 | $0.66 | ||||
| Claude Haiku 5.5 | 54.8 | $0.37 | |||
| 54.5 | $0.79 | ||||
| Grok 4.7 | 54.1 | $3.92 | |||
| MiMo-V2.6 Pro | 53.9 | $0.40 | |||
| GLM-5.3 | 52.3 | $8.77 | |||
| 52.2 | $0.68 | ||||
| 51.6 | $1.68 | ||||
| 51.6 | $0.43 | ||||
| 50.0 | $0.48 | ||||
| 46.1 | $0.88 |
3 models have held number one on the official leaderboard since its release in October 2026
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official Fiction Bench (Story Hacker) leaderboard
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official Fiction Bench (Story Hacker) leaderboard
No lab model card reports it yet
We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.
More in Writing and taste
- Design Bench (Moonshot) In use Models make graphic designs from direct prompts or few-shot reference images. An automated grader rates the results as Elo scores. Kimi K3.1 Pro 957
- WritingBench Fading Models complete 1,000 real-world writing requests across 6 domains and 100 subdomains. A judge model scores each reply on five query-specific criteria. GPT-5 83.9
- Creative Writing (MiniMax in-house) Only in system cards Models write creative pieces for real Hailuo AI user requests. Automatic assessments score the writing.
MiniMax-Text-01 81.3%
- DeepSeek Chinese functional writing evaluation Only in system cards Models answer 3,170 everyday Chinese writing prompts such as reports, emails and speeches. Pairwise judging against a baseline model gives a win rate.
DeepSeek-V4-Pro 62.6%
- Slide generation human evaluation (Z.ai) Only in system cards People compare HTML slide decks from two models on content, layout and visual quality. Metric is overall win rate. GLM-5 67.5%
- Task-dependent output homogenization evaluation Only in system cards Models answer prompts several times. A judge scores functional diversity where it is wanted and consistency where one answer is right.
Muse Spark 0.21
Data audit
All 10 checks pass. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 4 pass 1 note
- Pass Independent runs Scores from the official leaderboard (42 entries).
- Pass Models scored 36 models we track, plus 6 other entries on its leaderboard.
- Pass Newest score The newest is from November 2026, for GPT-6.2 Astra.
- Pass Frontier models 10 of the top 10 have a score, led by GPT-6.2 Astra (#1).
- Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
- Pass Importer We read the board on bench.storyhacker.ai: 36 rows for models we track and 6 other entries at the last read.
- Pass Last read November 20th, 2026.
- Pass Read worked The last read worked.
Consistency 1 pass 1 note
- Note Lab claims No model has both a lab number and an independent run.
- Pass Known issues No published audit found problems.