Rocinante 12B - NC Bench

thedrummer/rocinante-12b

Rocinante 12B

via OpenRouter

Release Date

Sep 30th, 2024

Context Size

32k

Reasoning

Benchmark Cost

$0.62

Speed

37.2 tok/s

Subcategories

Bad Writing Habits

Detects common prose quality anti-patterns in AI-generated creative writing, including passive voice, past progressive overuse, weak dialogue tags, filter words, purple prose, cliches, AI-ism words/adverbs/names, and more.

Scenario	#1	#2	#3	#4	#5	Total
Detailed Writing Rules
Fantasy: entering an ancient ruin Creative Writing Hallucination	89	80	78	77	70	79%
Horror: alone in an eerie place at night Creative Writing Hallucination	92	87	81	81	80	84%
Literary fiction: old friends reunite Creative Writing Hallucination	86	86	83	76	74	81%
Mystery: examining a crime scene Creative Writing Hallucination	88	88	88	88	82	87%
Romance: separated couple reunites Creative Writing Hallucination	90	84	80	77	77	82%
Thriller: chase through city streets Creative Writing Hallucination	92	89	87	86	81	87%
Detailed Writing Rules						83.23%
genre
Fantasy: entering an ancient ruin Creative Writing Hallucination	82	81	80	77	72	78%
Horror: alone in an eerie place at night Creative Writing Hallucination	87	83	83	79	72	81%
Literary fiction: old friends reunite Creative Writing Hallucination	86	83	78	75	74	79%
Mystery: examining a crime scene Creative Writing Hallucination	92	89	87	79	76	85%
Romance: separated couple reunites Creative Writing Hallucination	92	90	88	85	67	84%
Thriller: chase through city streets Creative Writing Hallucination	87	81	79	78	75	80%
genre						81.23%
Novelcrafter Default Prompt
Fantasy: entering an ancient ruin Creative Writing Hallucination	81	80	78	76	74	78%
Horror: alone in an eerie place at night Creative Writing Hallucination	90	86	85	79	79	84%
Literary fiction: old friends reunite Creative Writing Hallucination	88	80	79	75	73	79%
Mystery: examining a crime scene Creative Writing Hallucination	92	91	80	77	71	82%
Romance: separated couple reunites Creative Writing Hallucination	83	82	77	76	74	78%
Thriller: chase through city streets Creative Writing Hallucination	93	91	87	87	73	86%
Novelcrafter Default Prompt						81.26%
81.90%

Codex Extraction

Evaluates a model's ability to extract structured codex entries (characters, locations, objects, lore) from prose passages and return them as well-formed XML.

Scenario	#1	#2	#3	#4	#5	Total
Long: The Spire of Echoes (Dense) Tooling Reasoning	88	86	80	0	0	51%
Medium: The Hollow (Inferred) Tooling Reasoning	83	74	68	66	3	59%
Medium: Through the Thornveil (Scattered) Tooling Reasoning	82	72	68	46	0	54%
Short: The Rusty Lantern (Explicit) Tooling Reasoning	79	62	0	0	0	28%
47.95%

Codex Red Herring (False Positive Detection)

Tests whether models correctly report "no violations" when a codex is fully consistent with the prose passage. Models that hallucinate false violations (false positives) fail. Uses a 2×2 matrix of text length × codex size, with bare and detailed-entry variants.

Scenario	#1	#2	#3	#4	#5	#6	#7	#8	#9	#10	Total
basic entries
Long text (~1594 words), big codex (51 entries) Hallucination	100	100	100	35	35	25	13	10	5	3	43%
Long text (~1594 words), small codex (11 entries) Hallucination	100	33	31	17	10	8	7	6	4	3	22%
Short text (~524 words), big codex (51 entries) Hallucination	100	50	25	17	13	10	10	8	7	0	24%
Short text (~524 words), small codex (11 entries) Hallucination	100	100	100	50	10	10	10	6	5	4	39%
basic entries											31.98%
detailed entries
Long text (~1594 words), big codex (51 detailed entries) Hallucination	50	50	32	31	31	25	6	6	5	4	24%
Long text (~1594 words), small codex (11 detailed entries) Hallucination	30	13	10	7	7	6	4	4	4	3	9%
Short text (~524 words), big codex (51 detailed entries) Hallucination	33	31	17	13	10	8	6	6	6	4	13%
Short text (~524 words), small codex (11 detailed entries) Hallucination	100	100	100	100	25	10	6	4	4	3	45%
detailed entries											22.83%
27.40%

Codex Violation Detection

Detects factual inconsistencies between a story bible and prose passages. The model must output structured XML identifying each violation with paragraph number and substring.

Scenario	#1	#2	#3	#4	#5	#6	#7	#8	#9	#10	Total
matrix
Large codex (40 entries), long passage (1,019 words) Tooling Reasoning	44	44	42	33	33	33	33	33	3	0	30%
Large codex (40 entries), short passage (165 words) Tooling Reasoning	51	44	35	33	33	33	33	3	0	0	27%
Small codex (7 entries), long passage (734 words) Tooling Reasoning	70	65	44	38	33	33	33	30	0	0	35%
Small codex (7 entries), short passage (165 words) Tooling Reasoning	80	80	72	61	61	55	52	52	52	42	60%
matrix											37.99%
tiers
5 codex entries Tooling Reasoning	67	57	53	40	38	37	33	3	3	3	34%
10 codex entries Tooling Reasoning	72	72	64	64	51	33	33	3	0	0	39%
20 codex entries Tooling Reasoning	43	42	40	38	38	38	33	33	29	28	36%
40 codex entries Tooling Reasoning	44	42	38	33	33	33	33	17	3	0	28%
tiers											34.22%
36.11%

Data extraction

Extract key details from a given block of text.

Scenario	#1	#2	#3	#4	#5	#6	#7	#8	#9	#10	Total
All valid emails Utility	100	100	100	100	100	100	100	100	100	100	100%
Contextual pronoun Utility	100	100	100	100	100	100	100	50	50	50	85%
Fruits excluding citrus Utility	100	100	100	100	100	100	100	100	100	0	90%
Future event time Reasoning	100	50	50	50	50	50	50	50	50	50	55%
Guess the pet Reasoning	100	100	100	100	100	100	50	50	50	50	80%
Highest-rated movie Reasoning	100	100	100	100	100	100	50	50	0	0	70%
Indirect birth year Reasoning	100	100	100	100	100	50	50	50	50	0	70%
What instrument does Lucy play? Reasoning	100	100	100	50	50	50	50	50	50	50	65%
What's the color of the car? Utility	100	100	100	100	100	100	100	50	50	50	85%
What's the correct time? Reasoning	100	100	100	100	100	0	0	0	0	0	50%
Who's the sister? Reasoning	100	100	100	100	100	100	100	100	50	50	90%
Who's the tallest? Reasoning	100	100	100	100	100	100	100	100	100	50	95%
77.92%

Dialogue tags

Various tasks related to dialogue tags in text.

Scenario	#1	#2	#3	#4	#5	#6	#7	#8	#9	#10	Total
dialogue-200
Write 200 words with 10% dialogue Rule Following Utility	50	49	49	45	22	0	0	0	0	0	22%
Write 200 words with 50% dialogue Rule Following Utility	48	44	40	25	23	17	2	0	0	0	20%
Write 200 words with 90% dialogue Rule Following Utility	49	49	46	18	7	5	0	0	0	0	18%
dialogue-200											19.63%
dialogue-500
Write 500 words with 30% dialogue Rule Following Utility	43	3	0	0	0	0	0	0	0	0	5%
Write 500 words with 50% dialogue Rule Following Utility	100	33	24	7	3	1	0	0	0	0	17%
Write 500 words with 70% dialogue Rule Following Utility	91	49	38	15	6	0	0	0	0	0	20%
dialogue-500											13.79%
Ungrouped
Write unattributed dialogue Rule Following	100	100	100	100	61	14	14	1	0	0	49%
21.31%

Language Comprehension

Does the model understand more than just English?

Scenario	#1	#2	#3	#4	#5	Total
Asking for directions (Dutch) Language	100	100	0	0	0	40%
Asking for directions (German) Language	100	100	100	0	0	60%
Friend got new kittens (German) Language	100	100	100	100	100	100%
Friend got new kittens (Tagalog) Language	100	100	100	100	100	100%
75.00%

Language Writing

Can the model generate text in different languages?

Scenario	#1	#2	#3	#4	#5	Total
Character dialogue (French) in a story Language	100	94	67	63	50	75%
Character dialogue (German) in a story Language	100	100	80	67	0	69%
Character dialogue (Hindi) in a story Language	67	50	0	0	0	23%
Character dialogue (Italian) in a story Language	71	70	43	0	0	37%
Character dialogue (Spanish) in a story Language	64	64	63	54	33	55%
51.90%

N-Length Sentences

Write sentences with exactly N words

Scenario	#1	#2	#3	#4	#5	#6	#7	#8	#9	#10	Total
Write sentences with 5 words each Utility	88	78	72	66	62	59	59	40	34	22	58%
Write sentences with 10 words each Utility	56	43	35	31	26	21	14	3	1	0	23%
Write sentences with 20 words each Utility	0	0	0	0	0	0	0	0	0	0	0%
27.05%

Novel outline

Handle questions about the outline of a novel in various formats

Scenario	#1	#2	#3	#4	#5	#6	#7	#8	#9	#10	Total
outline-count
Count acts Utility	100	100	100	100	100	100	100	100	100	100	100%
Count chapters Utility	100	100	100	100	100	100	100	100	100	100	100%
Count scenes Utility	0	0	0	0	0	0	0	0	0	0	0%
outline-count											66.67%
pov-count
Count point of views for Jack and Olivia Utility	100	50	50	50	50	50	0	0	0	0	35%
Count point of views for Jack Harper Utility	100	100	100	100	0	0	0	0	0	0	40%
Count point of views for Olivia Utility	100	100	0	0	0	0	0	0	0	0	20%
pov-count											31.67%
49.17%

Text Replacement

Tests deterministic text transformations: renaming characters/locations, expanding contractions, tense rewriting, POV shifts, gender swaps, combined transformations, and word avoidance. Scored by checking each expected change independently.

Scenario	#1	#2	#3	#4	#5	#6	#7	Total
Generic Prompt
Avoid said/asked/replied/answered Text Editing	64	58	56	56	36	36	31	48%
Character rename: Elena->Mirabel, Gregor->Aldric Text Editing	100	100	100	100	95	94	87	97%
Combined: 3rd person past → 1st person present Text Editing	90	22	19	17	13	13	0	25%
Expand all contractions Text Editing	98	95	91	80	74	72	54	81%
Location rename: market square, outer ring, bridge, northern mines Text Editing	96	83	74	45	42	42	2	55%
Multi-character gender swap: Priya(F)->Rohan(M), Mara unchanged Text Editing	100	79	72	68	63	63	0	63%
Passive voice → active voice Text Editing Hallucination	37	36	26	23	23	15	12	24%
POV shift: 3rd person to 1st person (Elena's perspective) Text Editing	100	71	68	43	39	8	0	47%
Tense rewriting: past to present Text Editing	99	99	99	97	71	16	0	69%
Generic Prompt								56.51%
Specific Prompt
Avoid said/asked/replied/answered Text Editing	64	61	61	46	46	46	30	51%
Character rename: Elena->Mirabel, Gregor->Aldric Text Editing	100	100	100	100	100	100	95	99%
Combined: 3rd person past → 1st person present Text Editing	99	99	98	98	93	93	90	96%
Expand all contractions Text Editing	82	81	76	71	0	0	0	44%
Location rename: market square, outer ring, bridge, northern mines Text Editing	100	100	100	100	100	100	96	99%
Multi-character gender swap: Priya(F)->Rohan(M), Mara unchanged Text Editing	100	100	100	98	97	87	0	83%
Passive voice → active voice Text Editing Hallucination	78	73	73	53	43	27	20	52%
POV shift: 3rd person to 1st person (Elena's perspective) Text Editing	100	100	100	97	49	41	37	75%
Tense rewriting: past to present Text Editing	99	99	99	99	99	98	1	85%
Specific Prompt								76.02%
66.27%

Tool usage within Novelcrafter

Output messages that are related to tool usage within Novelcrafter

Scenario	#1	#2	#3	#4	#5	#6	#7	#8	#9	#10	Total
Create alternate prose sections Tooling	100	67	33	0	0	0	0	0	0	0	20%
20.00%

Voice/dialogue sheets

Extract dialogue from given text as voice sheets.

Scenario	#1	#2	#3	#4	#5	#6	#7	#8	#9	#10	Total
Multiple speakers Rule Following	0	0	0	0	0	0	0	0	0	0	0%
Simple Rule Following	100	100	100	100	100	100	0	0	0	0	60%
Simple (1-shot) 1-shot Rule Following	100	100	100	100	100	100	100	100	0	0	80%
Simple (5-shot) Few-shot Rule Following	100	100	100	100	100	100	100	0	0	0	70%
Unattributed dialogue Rule Following	100	100	100	100	100	0	0	0	0	0	50%
52.00%

Write N of X

Write exactly N words/sentences/paragraphs...

Scenario	#1	#2	#3	#4	#5	#6	#7	#8	#9	#10	Total
paragraphs
1 paragraph summary Utility	100	100	100	100	100	100	0	0	0	0	60%
3 paragraph summary Utility	100	100	100	0	0	0	0	0	0	0	30%
5 paragraph summary Utility	100	100	0	0	0	0	0	0	0	0	20%
paragraphs											36.67%
sentences
1 sentence summary Utility	100	100	100	100	100	98	98	92	0	0	79%
3 sentence summary Utility	100	100	100	100	100	100	98	98	77	2	88%
10 sentence summary Utility	100	100	92	92	77	27	2	0	0	0	49%
20 sentence summary Utility	100	100	92	54	54	0	0	0	0	0	40%
50 sentence summary Utility	92	9	0	0	0	0	0	0	0	0	10%
sentences											53.12%
words
10 word summary Utility	98	77	0	0	0	0	0	0	0	0	18%
20 word summary Utility	92	0	0	0	0	0	0	0	0	0	9%
50 word summary Utility	0	0	0	0	0	0	0	0	0	0	0%
100 word summary Utility	0	0	0	0	0	0	0	0	0	0	0%
200 word summary Utility	0	0	0	0	0	0	0	0	0	0	0%
words											5.36%
30.95%