Mistral Small 4 (Reasoning)

mistralai/mistral-small-2603

Mistral Small 4 (Reasoning)

via OpenRouter

Release Date

Mar 16th, 2026

Context Size

265k

Reasoning

Yes

Benchmark Cost

$1.29

Speed

144.6 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	80	76	76	75	72	76%
Horror: alone in an eerie place at night Creative Writing Hallucination	86	85	84	82	78	83%
Literary fiction: old friends reunite Creative Writing Hallucination	88	87	86	85	80	85%
Mystery: examining a crime scene Creative Writing Hallucination	88	83	82	82	81	83%
Romance: separated couple reunites Creative Writing Hallucination	85	84	83	83	81	83%
Thriller: chase through city streets Creative Writing Hallucination	86	84	83	82	77	82%
Detailed Writing Rules						82.13%
genre
Fantasy: entering an ancient ruin Creative Writing Hallucination	78	73	72	71	70	73%
Horror: alone in an eerie place at night Creative Writing Hallucination	85	81	80	79	73	80%
Literary fiction: old friends reunite Creative Writing Hallucination	87	82	80	79	77	81%
Mystery: examining a crime scene Creative Writing Hallucination	87	86	83	82	78	83%
Romance: separated couple reunites Creative Writing Hallucination	85	82	82	81	79	82%
Thriller: chase through city streets Creative Writing Hallucination	89	84	83	83	74	83%
genre						80.24%
Novelcrafter Default Prompt
Fantasy: entering an ancient ruin Creative Writing Hallucination	78	78	76	73	73	76%
Horror: alone in an eerie place at night Creative Writing Hallucination	85	85	84	79	77	82%
Literary fiction: old friends reunite Creative Writing Hallucination	90	87	85	83	80	85%
Mystery: examining a crime scene Creative Writing Hallucination	89	85	84	82	78	84%
Romance: separated couple reunites Creative Writing Hallucination	90	88	83	83	79	85%
Thriller: chase through city streets Creative Writing Hallucination	87	86	85	82	79	84%
Novelcrafter Default Prompt						82.44%
81.60%

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	89	87	90%
Medium: The Hollow (Inferred) Tooling Reasoning	94	93	90	90	90	91%
Medium: Through the Thornveil (Scattered) Tooling Reasoning	94	94	86	86	85	89%
Short: The Rusty Lantern (Explicit) Tooling Reasoning	96	94	92	88	86	91%
90.28%

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	100	100	100	100	100	100	17	92%
Long text (~1594 words), small codex (11 entries) Hallucination	100	100	100	100	100	100	25	25	25	17	69%
Short text (~524 words), big codex (51 entries) Hallucination	100	100	100	100	100	100	100	100	100	25	93%
Short text (~524 words), small codex (11 entries) Hallucination	100	100	100	100	100	100	100	100	100	100	100%
basic entries											88.33%
detailed entries
Long text (~1594 words), big codex (51 detailed entries) Hallucination	100	100	100	100	100	25	25	25	17	17	61%
Long text (~1594 words), small codex (11 detailed entries) Hallucination	100	100	100	100	100	100	100	25	25	25	78%
Short text (~524 words), big codex (51 detailed entries) Hallucination	100	100	100	100	100	100	100	100	100	25	93%
Short text (~524 words), small codex (11 detailed entries) Hallucination	100	100	100	100	100	100	100	25	17	17	76%
detailed entries											76.67%
82.50%

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	81	81	80	80	79	79	78	76	75	73	78%
Large codex (40 entries), short passage (165 words) Tooling Reasoning	93	91	90	89	88	88	88	81	79	78	87%
Small codex (7 entries), long passage (734 words) Tooling Reasoning	93	88	84	84	84	83	80	80	75	71	82%
Small codex (7 entries), short passage (165 words) Tooling Reasoning	97	97	93	93	93	88	88	88	84	83	90%
matrix											84.44%
tiers
5 codex entries Tooling Reasoning	100	100	93	93	93	93	91	91	91	83	93%
10 codex entries Tooling Reasoning	100	100	100	100	100	100	92	86	79	72	93%
20 codex entries Tooling Reasoning	94	91	88	86	86	82	82	82	79	73	84%
40 codex entries Tooling Reasoning	100	97	97	97	97	93	93	93	93	76	94%
tiers											90.98%
87.71%

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	100	100	50	50	50	50	50	50	50	50	60%
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	0	0	80%
Indirect birth year Reasoning	100	100	100	100	100	100	100	100	100	100	100%
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	100	100	100	100	0	0	0	0	0	0	40%
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%
90.00%

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	100	81	59	50	50	48	43	2	0	0	43%
Write 200 words with 50% dialogue Rule Following Utility	98	72	54	50	50	49	44	21	7	1	44%
Write 200 words with 90% dialogue Rule Following Utility	98	60	50	50	49	48	0	0	0	0	36%
dialogue-200											41.11%
dialogue-500
Write 500 words with 30% dialogue Rule Following Utility	18	12	3	0	0	0	0	0	0	0	3%
Write 500 words with 50% dialogue Rule Following Utility	53	50	2	0	0	0	0	0	0	0	11%
Write 500 words with 70% dialogue Rule Following Utility	48	45	24	20	19	16	2	0	0	0	17%
dialogue-500											10.38%
Ungrouped
Write unattributed dialogue Rule Following	100	100	100	100	100	100	100	100	100	100	100%
36.35%

Language Comprehension

Does the model understand more than just English?

Scenario	#1	#2	#3	#4	#5	Total
Asking for directions (Dutch) Language	0	0	0	0	0	0%
Asking for directions (German) Language	100	100	100	0	0	60%
Friend got new kittens (German) Language	100	100	0	0	0	40%
Friend got new kittens (Tagalog) Language	100	100	100	100	100	100%
50.00%

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	100	100	92	81	95%
Character dialogue (German) in a story Language	100	100	98	53	0	70%
Character dialogue (Hindi) in a story Language	100	100	91	50	50	78%
Character dialogue (Italian) in a story Language	100	100	100	73	0	75%
Character dialogue (Spanish) in a story Language	100	89	0	0	0	38%
71.06%

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	100	100	100	100	98	97	96	94	94	91	97%
Write sentences with 10 words each Utility	100	100	97	90	87	80	75	65	58	50	80%
Write sentences with 20 words each Utility	100	100	100	100	94	93	90	90	89	84	94%
90.42%

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	100	100	100	100	100	100	100	0	90%
outline-count											96.67%
pov-count
Count point of views for Jack and Olivia Utility	100	100	100	100	100	50	50	50	50	0	70%
Count point of views for Jack Harper Utility	100	100	100	100	100	100	100	0	0	0	70%
Count point of views for Olivia Utility	100	100	100	100	100	0	0	0	0	0	50%
pov-count											63.33%
80.00%

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	100	100	100	100	100	100%
Character rename: Elena->Mirabel, Gregor->Aldric Text Editing	100	100	100	100	100	100	100	100%
Combined: 3rd person past → 1st person present Text Editing	99	98	98	98	96	96	25	87%
Expand all contractions Text Editing	100	100	99	99	98	98	95	99%
Location rename: market square, outer ring, bridge, northern mines Text Editing	100	100	100	100	100	96	96	99%
Multi-character gender swap: Priya(F)->Rohan(M), Mara unchanged Text Editing	100	100	99	99	95	73	37	86%
Passive voice → active voice Text Editing Hallucination	93	91	91	90	86	79	78	87%
POV shift: 3rd person to 1st person (Elena's perspective) Text Editing	100	100	100	99	85	68	68	89%
Tense rewriting: past to present Text Editing	100	99	97	96	93	93	90	95%
Generic Prompt								93.50%
Specific Prompt
Avoid said/asked/replied/answered Text Editing	100	100	100	100	100	100	97	100%
Character rename: Elena->Mirabel, Gregor->Aldric Text Editing	100	100	100	100	100	100	100	100%
Combined: 3rd person past → 1st person present Text Editing	100	99	99	96	93	89	83	94%
Expand all contractions Text Editing	100	99	99	99	99	98	69	95%
Location rename: market square, outer ring, bridge, northern mines Text Editing	100	100	100	100	100	96	96	99%
Multi-character gender swap: Priya(F)->Rohan(M), Mara unchanged Text Editing	100	100	99	99	99	99	99	99%
Passive voice → active voice Text Editing Hallucination	93	93	93	92	90	89	40	84%
POV shift: 3rd person to 1st person (Elena's perspective) Text Editing	100	100	100	100	100	100	65	95%
Tense rewriting: past to present Text Editing	100	100	100	100	100	97	95	99%
Specific Prompt								96.09%
94.79%

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	100	100	100	0	0	0	0	0	0	0	30%
Simple (1-shot) 1-shot Rule Following	100	100	100	100	100	100	0	0	0	0	60%
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	100	100	100	0	0	80%
48.00%

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	100	100%
5 paragraph summary Utility	100	100	100	100	100	100	100	100	100	100	100%
paragraphs											100.00%
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	100	100	100	100	100	98	100%
20 sentence summary Utility	100	100	100	100	100	100	100	100	98	54	95%
50 sentence summary Utility	100	100	100	100	100	100	98	77	77	0	85%
sentences											96.06%
words
10 word summary Utility	100	100	100	100	100	100	100	100	98	98	100%
20 word summary Utility	100	100	100	100	100	100	100	100	98	98	100%
50 word summary Utility	100	100	100	100	100	54	27	27	2	0	61%
100 word summary Utility	100	92	77	27	27	9	9	2	0	0	34%
200 word summary Utility	0	0	0	0	0	0	0	0	0	0	0%
words											58.94%
82.69%