Roleplay combines public evaluations of emotional understanding, creative writing and story-related skills. It is a comparative reference, not a direct roleplay test.

Each benchmark is rescaled to 1–10 within its own leaderboard. We average results from the same publisher first, then give each publisher equal weight. Only verified model matches are included.

Creative Writing v3
claude-fable-5-1: 2,162.7
145 Models · 2026-10-25

LiveBench · Story constraint following
claude-fable-5-1-max-effort: 78.1
71 Models · 2026-10-25

LiveBench · Theory of mind
claude-fable-5-1-max-effort: 80.77
71 Models · 2026-10-25

Roleplay combines public evaluations of emotional understanding, creative writing and story-related skills. It is a comparative reference, not a direct roleplay test.

Each benchmark is rescaled to 1–10 within its own leaderboard. We average results from the same publisher first, then give each publisher equal weight. Only verified model matches are included.

LiveBench · Story constraint following
claude-haiku-5-5-xhigh-effort: 73.6
claude-haiku-5-5-max-effort: 64.53
71 Models · 2026-10-25

LiveBench · Theory of mind
claude-haiku-5-5-xhigh-effort: 76.92
claude-haiku-5-5-max-effort: 86.54
71 Models · 2026-10-25

Roleplay combines public evaluations of emotional understanding, creative writing and story-related skills. It is a comparative reference, not a direct roleplay test.

Each benchmark is rescaled to 1–10 within its own leaderboard. We average results from the same publisher first, then give each publisher equal weight. Only verified model matches are included.

Creative Writing v3
claude-opus-5: 2,133.5
145 Models · 2026-10-25

LiveBench · Story constraint following
claude-opus-5-max-effort: 61.32
71 Models · 2026-10-25

LiveBench · Theory of mind
claude-opus-5-max-effort: 78.85
71 Models · 2026-10-25

Roleplay combines public evaluations of emotional understanding, creative writing and story-related skills. It is a comparative reference, not a direct roleplay test.

Each benchmark is rescaled to 1–10 within its own leaderboard. We average results from the same publisher first, then give each publisher equal weight. Only verified model matches are included.

Creative Writing v3
claude-sonnet-5: 1,798
145 Models · 2026-10-25

LiveBench · Story constraint following
claude-sonnet-5-xhigh-effort: 62.32
71 Models · 2026-10-25

LiveBench · Theory of mind
claude-sonnet-5-xhigh-effort: 80.77
71 Models · 2026-10-25

Roleplay combines public evaluations of emotional understanding, creative writing and story-related skills. It is a comparative reference, not a direct roleplay test.

Each benchmark is rescaled to 1–10 within its own leaderboard. We average results from the same publisher first, then give each publisher equal weight. Only verified model matches are included.

LiveBench · Story constraint following
claude-opus-4-5-20251101-thinking-64k-high-effort: 65.85
71 Models · 2026-10-25

LiveBench · Theory of mind
claude-opus-4-5-20251101-thinking-64k-high-effort: 78.85
71 Models · 2026-10-25

WritingBench · Literature & arts
claude-opus-4-5-20251101: 81.41
54 Models · 2026-10-25

WritingBench · Character design
claude-opus-4-5-20251101: 85.6
54 Models · 2026-10-25

WritingBench · Plot development
claude-opus-4-5-20251101: 80.73
54 Models · 2026-10-25

WritingBench · Fan fiction
claude-opus-4-5-20251101: 81.25
54 Models · 2026-10-25

Scores from claude-haiku-4-5

Roleplay combines public evaluations of emotional understanding, creative writing and story-related skills. It is a comparative reference, not a direct roleplay test.

Each benchmark is rescaled to 1–10 within its own leaderboard. We average results from the same publisher first, then give each publisher equal weight. Only verified model matches are included.

LiveBench · Story constraint following (2025-12-23)
claude-haiku-4-5-20251001: 16.42
claude-haiku-4-5-20251001-thinking-64k: 57.27
65 Models · 2026-10-25

LiveBench · Theory of mind (2025-12-23)
claude-haiku-4-5-20251001: 50
claude-haiku-4-5-20251001-thinking-64k: 44.23
65 Models · 2026-10-25

WritingBench · Literature & arts
claude-haiku-4-5-20251001: 76.35
54 Models · 2026-10-25

WritingBench · Character design
claude-haiku-4-5-20251001: 79.2
54 Models · 2026-10-25

WritingBench · Plot development
claude-haiku-4-5-20251001: 77.82
54 Models · 2026-10-25

WritingBench · Fan fiction
claude-haiku-4-5-20251001: 77.5
54 Models · 2026-10-25

Scores from claude-opus-4-6

Roleplay combines public evaluations of emotional understanding, creative writing and story-related skills. It is a comparative reference, not a direct roleplay test.

Each benchmark is rescaled to 1–10 within its own leaderboard. We average results from the same publisher first, then give each publisher equal weight. Only verified model matches are included.

EQ-Bench 3
claude-opus-4-6: 1,712.4
83 Models · 2026-10-25

Creative Writing v3
claude-opus-4-6: 1,810.4
145 Models · 2026-10-25

LiveBench · Story constraint following
claude-opus-4-6-thinking-auto-high-effort: 66.13
71 Models · 2026-10-25

LiveBench · Theory of mind
claude-opus-4-6-thinking-auto-high-effort: 82.69
71 Models · 2026-10-25

Roleplay combines public evaluations of emotional understanding, creative writing and story-related skills. It is a comparative reference, not a direct roleplay test.

Each benchmark is rescaled to 1–10 within its own leaderboard. We average results from the same publisher first, then give each publisher equal weight. Only verified model matches are included.

EQ-Bench 3
claude-sonnet-4-6: 1,709.8
83 Models · 2026-10-25

Creative Writing v3
claude-sonnet-4-6: 1,811.2
145 Models · 2026-10-25

LiveBench · Story constraint following
claude-sonnet-4-6-thinking-auto-medium-effort: 68.9
71 Models · 2026-10-25

LiveBench · Theory of mind
claude-sonnet-4-6-thinking-auto-medium-effort: 73.08
71 Models · 2026-10-25

Scores from claude-opus-4-8

Roleplay combines public evaluations of emotional understanding, creative writing and story-related skills. It is a comparative reference, not a direct roleplay test.

Each benchmark is rescaled to 1–10 within its own leaderboard. We average results from the same publisher first, then give each publisher equal weight. Only verified model matches are included.

EQ-Bench 3
claude-opus-4-8: 2,057.8
83 Models · 2026-10-25

Creative Writing v3
claude-opus-4-8: 1,840.2
145 Models · 2026-10-25

LiveBench · Story constraint following
claude-opus-4-8-max-effort: 80.6
71 Models · 2026-10-25

LiveBench · Theory of mind
claude-opus-4-8-max-effort: 80.77
71 Models · 2026-10-25

Roleplay combines public evaluations of emotional understanding, creative writing and story-related skills. It is a comparative reference, not a direct roleplay test.

Each benchmark is rescaled to 1–10 within its own leaderboard. We average results from the same publisher first, then give each publisher equal weight. Only verified model matches are included.

LiveBench · Story constraint following
claude-sonnet-5-5-max-effort: 69.92
claude-sonnet-5-5-xhigh-effort: 77.08
71 Models · 2026-10-25

LiveBench · Theory of mind
claude-sonnet-5-5-max-effort: 86.54
claude-sonnet-5-5-xhigh-effort: 69.23
71 Models · 2026-10-25

Roleplay combines public evaluations of emotional understanding, creative writing and story-related skills. It is a comparative reference, not a direct roleplay test.

Each benchmark is rescaled to 1–10 within its own leaderboard. We average results from the same publisher first, then give each publisher equal weight. Only verified model matches are included.

EQ-Bench 3
claude-opus-4-6: 1,712.4
83 Models · 2026-10-25

Creative Writing v3
claude-opus-4-6: 1,810.4
145 Models · 2026-10-25

LiveBench · Story constraint following
claude-opus-4-6-thinking-auto-high-effort: 66.13
71 Models · 2026-10-25

LiveBench · Theory of mind
claude-opus-4-6-thinking-auto-high-effort: 82.69
71 Models · 2026-10-25

Roleplay combines public evaluations of emotional understanding, creative writing and story-related skills. It is a comparative reference, not a direct roleplay test.

Each benchmark is rescaled to 1–10 within its own leaderboard. We average results from the same publisher first, then give each publisher equal weight. Only verified model matches are included.

EQ-Bench 3
claude-opus-4-8: 2,057.8
83 Models · 2026-10-25

Creative Writing v3
claude-opus-4-8: 1,840.2
145 Models · 2026-10-25

LiveBench · Story constraint following
claude-opus-4-8-max-effort: 80.6
71 Models · 2026-10-25

LiveBench · Theory of mind
claude-opus-4-8-max-effort: 80.77
71 Models · 2026-10-25

Scores from claude-opus-4-5

Roleplay combines public evaluations of emotional understanding, creative writing and story-related skills. It is a comparative reference, not a direct roleplay test.

Each benchmark is rescaled to 1–10 within its own leaderboard. We average results from the same publisher first, then give each publisher equal weight. Only verified model matches are included.

LiveBench · Story constraint following
claude-opus-4-5-20251101-thinking-64k-high-effort: 65.85
71 Models · 2026-10-25

LiveBench · Theory of mind
claude-opus-4-5-20251101-thinking-64k-high-effort: 78.85
71 Models · 2026-10-25

WritingBench · Literature & arts
claude-opus-4-5-20251101: 81.41
54 Models · 2026-10-25

WritingBench · Character design
claude-opus-4-5-20251101: 85.6
54 Models · 2026-10-25

WritingBench · Plot development
claude-opus-4-5-20251101: 80.73
54 Models · 2026-10-25

WritingBench · Fan fiction
claude-opus-4-5-20251101: 81.25
54 Models · 2026-10-25

Roleplay combines public evaluations of emotional understanding, creative writing and story-related skills. It is a comparative reference, not a direct roleplay test.

Each benchmark is rescaled to 1–10 within its own leaderboard. We average results from the same publisher first, then give each publisher equal weight. Only verified model matches are included.

EQ-Bench 3
claude-opus-4-7: 1,914
83 Models · 2026-10-25

Creative Writing v3
claude-opus-4-7: 1,915
145 Models · 2026-10-25

LiveBench · Story constraint following
claude-opus-4-7-xhigh-effort: 69.4
71 Models · 2026-10-25

LiveBench · Theory of mind
claude-opus-4-7-xhigh-effort: 80.77
71 Models · 2026-10-25

Roleplay combines public evaluations of emotional understanding, creative writing and story-related skills. It is a comparative reference, not a direct roleplay test.

Each benchmark is rescaled to 1–10 within its own leaderboard. We average results from the same publisher first, then give each publisher equal weight. Only verified model matches are included.

EQ-Bench 3
claude-fable-5: 2,096.4
83 Models · 2026-10-25

Creative Writing v3
claude-fable-5: 1,943.6
145 Models · 2026-10-25

LiveBench · Story constraint following
claude-fable-5-max-effort: 71.7
71 Models · 2026-10-25

LiveBench · Theory of mind
claude-fable-5-max-effort: 84.62
71 Models · 2026-10-25

Scores from claude-opus-5

Roleplay combines public evaluations of emotional understanding, creative writing and story-related skills. It is a comparative reference, not a direct roleplay test.

Each benchmark is rescaled to 1–10 within its own leaderboard. We average results from the same publisher first, then give each publisher equal weight. Only verified model matches are included.

Creative Writing v3
claude-opus-5: 2,133.5
145 Models · 2026-10-25

LiveBench · Story constraint following
claude-opus-5-max-effort: 61.32
71 Models · 2026-10-25

LiveBench · Theory of mind
claude-opus-5-max-effort: 78.85
71 Models · 2026-10-25

Roleplay combines public evaluations of emotional understanding, creative writing and story-related skills. It is a comparative reference, not a direct roleplay test.

Each benchmark is rescaled to 1–10 within its own leaderboard. We average results from the same publisher first, then give each publisher equal weight. Only verified model matches are included.

LiveBench · Story constraint following (2025-12-23)
claude-haiku-4-5-20251001: 16.42
claude-haiku-4-5-20251001-thinking-64k: 57.27
65 Models · 2026-10-25

LiveBench · Theory of mind (2025-12-23)
claude-haiku-4-5-20251001: 50
claude-haiku-4-5-20251001-thinking-64k: 44.23
65 Models · 2026-10-25

WritingBench · Literature & arts
claude-haiku-4-5-20251001: 76.35
54 Models · 2026-10-25

WritingBench · Character design
claude-haiku-4-5-20251001: 79.2
54 Models · 2026-10-25

WritingBench · Plot development
claude-haiku-4-5-20251001: 77.82
54 Models · 2026-10-25

WritingBench · Fan fiction
claude-haiku-4-5-20251001: 77.5
54 Models · 2026-10-25

Scores from claude-sonnet-5

Roleplay combines public evaluations of emotional understanding, creative writing and story-related skills. It is a comparative reference, not a direct roleplay test.

Each benchmark is rescaled to 1–10 within its own leaderboard. We average results from the same publisher first, then give each publisher equal weight. Only verified model matches are included.

Creative Writing v3
claude-sonnet-5: 1,798
145 Models · 2026-10-25

LiveBench · Story constraint following
claude-sonnet-5-xhigh-effort: 62.32
71 Models · 2026-10-25

LiveBench · Theory of mind
claude-sonnet-5-xhigh-effort: 80.77
71 Models · 2026-10-25

Scores from claude-sonnet-4-6

Roleplay combines public evaluations of emotional understanding, creative writing and story-related skills. It is a comparative reference, not a direct roleplay test.

Each benchmark is rescaled to 1–10 within its own leaderboard. We average results from the same publisher first, then give each publisher equal weight. Only verified model matches are included.

EQ-Bench 3
claude-sonnet-4-6: 1,709.8
83 Models · 2026-10-25

Creative Writing v3
claude-sonnet-4-6: 1,811.2
145 Models · 2026-10-25

LiveBench · Story constraint following
claude-sonnet-4-6-thinking-auto-medium-effort: 68.9
71 Models · 2026-10-25

LiveBench · Theory of mind
claude-sonnet-4-6-thinking-auto-medium-effort: 73.08
71 Models · 2026-10-25

Roleplay combines public evaluations of emotional understanding, creative writing and story-related skills. It is a comparative reference, not a direct roleplay test.

Each benchmark is rescaled to 1–10 within its own leaderboard. We average results from the same publisher first, then give each publisher equal weight. Only verified model matches are included.

Creative Writing v3
claude-opus-5-5: 2,050.8
145 Models · 2026-10-25

LiveBench · Story constraint following
claude-opus-5-5-max-effort: 70.92
claude-opus-5-5-xhigh-effort: 71.18
71 Models · 2026-10-25

LiveBench · Theory of mind
claude-opus-5-5-max-effort: 84.62
claude-opus-5-5-xhigh-effort: 84.62
71 Models · 2026-10-25

Scores from claude-opus-4-7

Roleplay combines public evaluations of emotional understanding, creative writing and story-related skills. It is a comparative reference, not a direct roleplay test.

Each benchmark is rescaled to 1–10 within its own leaderboard. We average results from the same publisher first, then give each publisher equal weight. Only verified model matches are included.

EQ-Bench 3
claude-opus-4-7: 1,914
83 Models · 2026-10-25

Creative Writing v3
claude-opus-4-7: 1,915
145 Models · 2026-10-25

LiveBench · Story constraint following
claude-opus-4-7-xhigh-effort: 69.4
71 Models · 2026-10-25

LiveBench · Theory of mind
claude-opus-4-7-xhigh-effort: 80.77
71 Models · 2026-10-25

Roleplay combines public evaluations of emotional understanding, creative writing and story-related skills. It is a comparative reference, not a direct roleplay test.

Each benchmark is rescaled to 1–10 within its own leaderboard. We average results from the same publisher first, then give each publisher equal weight. Only verified model matches are included.

EQ-Bench 3
claude-sonnet-4.5: 1,534.6
83 Models · 2026-10-25

Creative Writing v3
claude-sonnet-4.5: 1,677.2
145 Models · 2026-10-25

LiveBench · Story constraint following (2025-12-23)
claude-sonnet-4-5-20250929: 28.18
claude-sonnet-4-5-20250929-thinking-64k: 58.4
65 Models · 2026-10-25

LiveBench · Theory of mind (2025-12-23)
claude-sonnet-4-5-20250929: 51.92
claude-sonnet-4-5-20250929-thinking-64k: 78.85
65 Models · 2026-10-25

WritingBench · Literature & arts
claude-sonnet-4-5-20250929: 81.26
54 Models · 2026-10-25

WritingBench · Character design
claude-sonnet-4-5-20250929: 85.2
54 Models · 2026-10-25

WritingBench · Plot development
claude-sonnet-4-5-20250929: 78.73
54 Models · 2026-10-25

WritingBench · Fan fiction
claude-sonnet-4-5-20250929: 81.5
54 Models · 2026-10-25

Scores from claude-sonnet-4-5

Roleplay combines public evaluations of emotional understanding, creative writing and story-related skills. It is a comparative reference, not a direct roleplay test.

Each benchmark is rescaled to 1–10 within its own leaderboard. We average results from the same publisher first, then give each publisher equal weight. Only verified model matches are included.

EQ-Bench 3
claude-sonnet-4.5: 1,534.6
83 Models · 2026-10-25

Creative Writing v3
claude-sonnet-4.5: 1,677.2
145 Models · 2026-10-25

LiveBench · Story constraint following (2025-12-23)
claude-sonnet-4-5-20250929: 28.18
claude-sonnet-4-5-20250929-thinking-64k: 58.4
65 Models · 2026-10-25

LiveBench · Theory of mind (2025-12-23)
claude-sonnet-4-5-20250929: 51.92
claude-sonnet-4-5-20250929-thinking-64k: 78.85
65 Models · 2026-10-25

WritingBench · Literature & arts
claude-sonnet-4-5-20250929: 81.26
54 Models · 2026-10-25

WritingBench · Character design
claude-sonnet-4-5-20250929: 85.2
54 Models · 2026-10-25

WritingBench · Plot development
claude-sonnet-4-5-20250929: 78.73
54 Models · 2026-10-25

WritingBench · Fan fiction
claude-sonnet-4-5-20250929: 81.5
54 Models · 2026-10-25

Scores from claude-opus-4-8

Roleplay combines public evaluations of emotional understanding, creative writing and story-related skills. It is a comparative reference, not a direct roleplay test.

Each benchmark is rescaled to 1–10 within its own leaderboard. We average results from the same publisher first, then give each publisher equal weight. Only verified model matches are included.

EQ-Bench 3
claude-opus-4-8: 2,057.8
83 Models · 2026-10-25

Creative Writing v3
claude-opus-4-8: 1,840.2
145 Models · 2026-10-25

LiveBench · Story constraint following
claude-opus-4-8-max-effort: 80.6
71 Models · 2026-10-25

LiveBench · Theory of mind
claude-opus-4-8-max-effort: 80.77
71 Models · 2026-10-25