Claude 3 Haiku - NC Bench

anthropic/claude-3-haiku

Claude 3 Haiku

via OpenRouter

Release Date

Mar 13th, 2024

Context Size

200k

Reasoning

Benchmark Cost

$1.27

Speed

1382.0 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	78	75	74	74	73	75%
Horror: alone in an eerie place at night Creative Writing Hallucination	81	79	71	66	64	72%
Literary fiction: old friends reunite Creative Writing Hallucination	90	78	78	77	77	80%
Mystery: examining a crime scene Creative Writing Hallucination	80	79	79	74	73	77%
Romance: separated couple reunites Creative Writing Hallucination	84	83	77	76	76	79%
Thriller: chase through city streets Creative Writing Hallucination	85	83	77	77	74	79%
Detailed Writing Rules						77.11%
genre
Fantasy: entering an ancient ruin Creative Writing Hallucination	78	77	75	72	69	74%
Horror: alone in an eerie place at night Creative Writing Hallucination	80	78	75	73	71	75%
Literary fiction: old friends reunite Creative Writing Hallucination	76	75	73	72	71	73%
Mystery: examining a crime scene Creative Writing Hallucination	85	77	75	73	72	76%
Romance: separated couple reunites Creative Writing Hallucination	80	75	73	71	71	74%
Thriller: chase through city streets Creative Writing Hallucination	84	79	74	73	68	76%
genre						74.82%
Novelcrafter Default Prompt
Fantasy: entering an ancient ruin Creative Writing Hallucination	77	72	72	70	68	72%
Horror: alone in an eerie place at night Creative Writing Hallucination	77	76	72	71	70	73%
Literary fiction: old friends reunite Creative Writing Hallucination	73	72	71	66	64	69%
Mystery: examining a crime scene Creative Writing Hallucination	83	80	80	75	75	79%
Romance: separated couple reunites Creative Writing Hallucination	77	74	71	68	67	71%
Thriller: chase through city streets Creative Writing Hallucination	84	76	75	75	75	77%
Novelcrafter Default Prompt						73.61%
75.18%

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	92	90	90	90	89	90%
Medium: The Hollow (Inferred) Tooling Reasoning	94	92	90	89	89	91%
Medium: Through the Thornveil (Scattered) Tooling Reasoning	93	91	87	85	82	88%
Short: The Rusty Lantern (Explicit) Tooling Reasoning	94	92	89	87	81	88%
89.24%

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	38	38	38	38	38	38	35	13	13	6	29%
Long text (~1594 words), small codex (11 entries) Hallucination	33	10	10	8	8	7	7	7	6	6	10%
Short text (~524 words), big codex (51 entries) Hallucination	33	33	32	10	7	7	7	7	6	5	15%
Short text (~524 words), small codex (11 entries) Hallucination	38	38	35	33	13	10	8	7	7	6	19%
basic entries											18.40%
detailed entries
Long text (~1594 words), big codex (51 detailed entries) Hallucination	42	38	33	13	13	13	6	6	6	6	17%
Long text (~1594 words), small codex (11 detailed entries) Hallucination	38	38	33	33	13	13	10	8	6	5	20%
Short text (~524 words), big codex (51 detailed entries) Hallucination	33	32	17	10	10	8	7	7	7	6	14%
Short text (~524 words), small codex (11 detailed entries) Hallucination	42	42	42	38	38	38	35	35	35	10	35%
detailed entries											21.48%
19.94%

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	61	58	51	49	47	46	44	44	44	44	49%
Large codex (40 entries), short passage (165 words) Tooling Reasoning	67	60	54	51	49	49	49	47	39	33	50%
Small codex (7 entries), long passage (734 words) Tooling Reasoning	77	77	73	73	73	73	73	73	61	52	71%
Small codex (7 entries), short passage (165 words) Tooling Reasoning	76	76	76	72	72	72	72	70	60	55	70%
matrix											59.84%
tiers
5 codex entries Tooling Reasoning	76	76	67	67	67	60	57	57	50	50	62%
10 codex entries Tooling Reasoning	79	79	79	79	66	64	58	58	58	58	68%
20 codex entries Tooling Reasoning	75	65	60	59	55	55	51	51	48	40	56%
40 codex entries Tooling Reasoning	44	44	44	42	42	42	40	40	40	39	42%
tiers											57.02%
58.43%

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	100	100	100	100%
Fruits excluding citrus Utility	100	100	100	100	100	100	100	100	100	100	100%
Future event time Reasoning	50	50	50	50	50	50	50	50	50	50	50%
Guess the pet Reasoning	100	100	100	100	100	100	100	100	100	100	100%
Highest-rated movie Reasoning	100	100	100	100	100	100	100	100	100	100	100%
Indirect birth year Reasoning	100	100	100	100	100	100	100	100	100	50	95%
What instrument does Lucy play? Reasoning	100	100	100	100	100	100	100	100	100	100	100%
What's the color of the car? Utility	100	100	100	100	100	100	100	100	100	100	100%
What's the correct time? Reasoning	0	0	0	0	0	0	0	0	0	0	0%
Who's the sister? Reasoning	100	100	100	100	100	100	100	100	100	100	100%
Who's the tallest? Reasoning	100	100	100	100	100	100	100	100	100	100	100%
87.08%

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	49	49	49	49	48	47	43	22	5	0	36%
Write 200 words with 50% dialogue Rule Following Utility	74	71	61	51	49	41	41	30	14	10	44%
Write 200 words with 90% dialogue Rule Following Utility	50	50	48	47	38	22	18	7	1	0	28%
dialogue-200											36.03%
dialogue-500
Write 500 words with 30% dialogue Rule Following Utility	79	66	15	3	1	0	0	0	0	0	17%
Write 500 words with 50% dialogue Rule Following Utility	34	2	1	0	0	0	0	0	0	0	4%
Write 500 words with 70% dialogue Rule Following Utility	49	44	30	25	22	14	13	1	0	0	20%
dialogue-500											13.39%
Ungrouped
Write unattributed dialogue Rule Following	100	100	100	100	100	100	100	100	100	100	100%
35.46%

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	100	100	100%
Friend got new kittens (German) Language	100	0	0	0	0	20%
Friend got new kittens (Tagalog) Language	100	100	100	100	100	100%
65.00%

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	50	50	50	43	59%
Character dialogue (German) in a story Language	100	100	50	50	43	69%
Character dialogue (Hindi) in a story Language	100	100	100	100	100	100%
Character dialogue (Italian) in a story Language	100	100	100	50	40	78%
Character dialogue (Spanish) in a story Language	100	100	100	100	88	98%
80.53%

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	89	85	81	78	72	72	68	65	47	1	66%
Write sentences with 10 words each Utility	97	84	83	82	78	75	74	63	61	31	73%
Write sentences with 20 words each Utility	27	15	13	10	4	0	0	0	0	0	7%
48.51%

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	100	100	0	0	0	0	0	0	0	0	20%
outline-count											73.33%
pov-count
Count point of views for Jack and Olivia Utility	100	100	100	50	0	0	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	0	0	0	0	0	0	0	0	0	10%
pov-count											28.33%
50.83%

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	100	100	64	50	48	22	22	58%
Character rename: Elena->Mirabel, Gregor->Aldric Text Editing	100	100	0	0	0	0	0	29%
Combined: 3rd person past → 1st person present Text Editing	93	92	0	0	0	0	0	26%
Expand all contractions Text Editing	98	98	0	0	0	0	0	28%
Location rename: market square, outer ring, bridge, northern mines Text Editing	0	0	0	0	0	0	0	0%
Multi-character gender swap: Priya(F)->Rohan(M), Mara unchanged Text Editing	86	86	86	86	86	86	78	85%
Passive voice → active voice Text Editing Hallucination	18	18	18	18	18	18	18	18%
POV shift: 3rd person to 1st person (Elena's perspective) Text Editing	0	0	0	0	0	0	0	0%
Tense rewriting: past to present Text Editing	0	0	0	0	0	0	0	0%
Generic Prompt								27.11%
Specific Prompt
Avoid said/asked/replied/answered Text Editing	100	100	100	100	75	69	53	85%
Character rename: Elena->Mirabel, Gregor->Aldric Text Editing	100	100	100	100	100	100	33	90%
Combined: 3rd person past → 1st person present Text Editing	94	94	93	93	93	93	93	94%
Expand all contractions Text Editing	100	100	100	100	100	99	99	100%
Location rename: market square, outer ring, bridge, northern mines Text Editing	100	100	100	100	100	100	100	100%
Multi-character gender swap: Priya(F)->Rohan(M), Mara unchanged Text Editing	100	100	100	100	100	100	100	100%
Passive voice → active voice Text Editing Hallucination	93	91	89	89	88	88	86	89%
POV shift: 3rd person to 1st person (Elena's perspective) Text Editing	97	97	97	97	97	97	97	97%
Tense rewriting: past to present Text Editing	100	100	100	100	100	100	100	100%
Specific Prompt								95.02%
61.07%

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	100	100	100	100	100	100	100	100	100	100%
100.00%

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	0	0	0	0	0	0	0	0	0	0	0%
Simple (1-shot) 1-shot Rule Following	100	100	100	100	100	100	100	100	100	100	100%
Simple (5-shot) Few-shot Rule Following	100	100	100	100	100	100	100	100	100	100	100%
Unattributed dialogue Rule Following	0	0	0	0	0	0	0	0	0	0	0%
40.00%

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	100	100	100	100	100%
3 paragraph summary Utility	100	100	100	100	100	100	100	100	100	0	90%
5 paragraph summary Utility	100	100	100	100	100	100	100	100	100	100	100%
paragraphs											96.67%
sentences
1 sentence summary Utility	100	100	100	100	100	100	100	100	100	100	100%
3 sentence summary Utility	100	100	100	100	100	100	100	100	100	100	100%
10 sentence summary Utility	100	100	100	100	77	77	77	54	54	27	77%
20 sentence summary Utility	100	100	100	0	0	0	0	0	0	0	30%
50 sentence summary Utility	27	0	0	0	0	0	0	0	0	0	3%
sentences											61.88%
words
10 word summary Utility	100	100	100	98	98	98	98	92	92	77	96%
20 word summary Utility	100	100	100	100	98	77	54	54	27	9	72%
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											33.49%
58.99%