NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-20-01KHY8B4PVPPMPSJC88NNBX066
Run Details
Model:
Ministral 3 8B
Test:
Language Writing
Scenario:
Character dialogue (Italian) in a story
Duration:
5.5s
Cost:
$0.000197
Provider:
Mistral
Evaluation summary
Score
Evaluator
Details Link
100.00%
Parse dialogue
(details)
Model input
System
You are an expert novel author in the cozy mystery genre. Take the following details about characters/locations/... into account: <character name="Isabella"> Isabella is a 30-year old college student on a study abroad program. Born in Italy, she is fluent in Italian and does not yet know any English. She enjoys landscape photography and is always seen with her camera. </character> <character name="Hunter"> Hunter is a 35-year old detective who is investigating a series of mysterious disappearances in the small town of Alborn, MN. He is a former marine and has a no-nonsense attitude. Hunter has no knowledge of Italian, not even basic phrases. </character> <location name="Alborn, MN"> Alborne is a small town in the Midwest with a population of 3,000. It is known for its annual apple festival. </location>
User/Human
Write the following scene: Isabella is sitting at a cafe, looking at her camera. Hunter approaches her and asks her about a recent event in the town. She responds troubled and shaken in Italian, and Hunter looks confused. They talk for a while, until Hunter gets a call and has to leave. Return the dialogue between Isabella and Hunter as a voice sheet: Isabella: "..." Hunter: "..."
Model response
Messages
Meta
AI
**VOICE SHEET: SCENE – THE CAFÉ CONFRONTATION** **Setting:** A quaint, sunlit café in Alborn, MN. The scent of coffee and cinnamon lingers in the air. Isabella sits at a small wooden table, her camera open beside her, her fingers tracing the lens thoughtfully. Hunter, in a rumpled jacket, leans against the table across from her, arms crossed. --- ### **DIALOGUE (Translated for Context – Hunter’s Perspective)** *(Hunter clears his throat, shifting uncomfortably. Isabella doesn’t look up from her camera.)* **Hunter:** *(in English, blunt)* "You’re the new girl, right? Study abroad?" *(Isabella flinches. She closes the camera with a sharp *click*. Her hands tremble slightly as she sets it down. She speaks rapidly in Italian, her voice tight with emotion.)* **Isabella:** *(distraught, urgent)* *"No… no, I’m not new. I’ve been here for weeks. And I don’t understand why you’re asking me these things—why you’re looking at me like I did something wrong!"* *(Hunter blinks, confused. He doesn’t speak Italian, so he just stares at her, hands raised in a gesture of surrender.)* **Hunter:** *(frowning)* "I’m not accusing you of anything. I’m just trying to figure out what happened to the missing people." **Isabella:** *(voice breaking, gesturing wildly with her hands)* *"You don’t understand! It wasn’t just *people*—it was *photos*! My photos! They were taken—from my camera! And now… now they’re gone!"* *(Hunter’s jaw tightens. He leans forward, suddenly alert.)* **Hunter:** "What do you mean, ‘taken’?" **Isabella:** *(whispering, horrified)* *"They were *stolen*. From my bag. From my room. And then… then they appeared *elsewhere*. In places I’ve never been. In photos I didn’t take!"* *(Hunter exhales sharply, rubbing his temples. He pulls out a notepad and pen, scribbling.)* **Hunter:** "Where were these photos found?" **Isabella:** *(voice trembling, pointing vaguely)* *"In the woods. Near the old mill. And… and in the church basement! Someone was *using* them!"* *(Hunter’s eyes flicker with recognition. He knows the mill and the church—both have been tied to the disappearances.)* **Hunter:** "Did you report this to the police?" **Isabella:** *(bitter, shaking her head)* *"The police? No! They don’t speak my language! And they don’t *care*! They just… they just tell me to ‘watch out for strangers’!"* *(Hunter’s expression darkens. He pulls out his phone, checking a message. His face tightens.)* **Hunter:** *(muttering, to himself)* "Shit." *(Isabella’s head snaps up, alarmed.)* **Isabella:** *(in Italian, frantic)* *"What? What is it?!" **Hunter:** *(without looking at her, already standing)* *"I’ve got to go. But—" *(he pulls a business card from his pocket, scribbles something on the back)* "—call me if you see or hear anything else. *Anything.*" *(He slides the card across the table.)* *(Isabella stares at it, then at him, her eyes wide with fear.)* **Isabella:** *(whispering, desperate)* *"You… you think they’ll come back?"* *(Hunter hesitates, then meets her gaze—serious, grim.)* **Hunter:** "I think someone’s playing a game. And I’m not letting them win." *(He turns and strides toward the door, his voice echoing as he calls over his shoulder—)* **Hunter:** "And *Isabella*?" *(She flinches at her name in English.)* "If you see a man in a dark coat with a camera… *run.*" *(The door swings shut behind him. Isabella sits frozen, the card trembling in her grip, her breath coming fast. Outside, the sound of Hunter’s boots fading into the distance.)* --- **END SCENE.** *(Note: For full immersion, Isabella’s Italian dialogue would be more fluid and emotional, with rapid-fire phrases like *"Non capisci! Non è giusto!"* ("You don’t understand! It’s not fair!") and *"Qualcuno sta usando le mie foto come un’arma!"* ("Someone is using my photos like a weapon!"). Hunter’s confusion would be palpable—he’d likely mutter under his breath in English, *"What the hell is she talking about?"* before realizing her urgency.)*
created_at
"2026-02-20T19:27:34.633Z"
2/20/2026, 7:27:34 PM
model
"mistralai/ministral-8b-2512"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
373
moderation_latency
(null)
generation_time
5196
tokens_prompt
321
tokens_completion
1101
native_tokens_prompt
277
native_tokens_completion
1037
native_tokens_completion_images
(null)
native_tokens_reasoning
0
native_tokens_cached
0
num_media_prompt
(null)
num_input_audio_prompt
(null)
num_media_completion
0
num_search_results
(null)
origin
"https://ncbench.com/"
is_byok
false
finish_reason
"stop"
native_finish_reason
"stop"
usage
0.0001971
router
(null)
provider_responses
0
endpoint_id
"c85d26e1-1e0c-449b-9775-2afce7ae510b"
id
"ce9a0fa3e71b42bf9daea2011482314e"
is_byok
false
latency
373
model_permaslug
"mistralai/ministral-8b-2512"
provider_name
"Mistral"
status
200
api_type
"completions"
id
"gen-1771615654-fWmFLmsmOOYSmtOh6Zkd"
upstream_id
"ce9a0fa3e71b42bf9daea2011482314e"
total_cost
0.0001971
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Mistral"
Evaluation details
Result
Evaluator
Details
Meta Data
100.00%
Parse dialogue
n/a
dialogue
0
name
"Hunter"
dialogue
"You’re the new girl, right? Study abroad?"
detectedLang
"en"
heavyLang
"en"
scores
en
0.8104864181933038
tl
0.637243047158404
no
0.5033112582781457
da
0.47735191637630664
fr
0.47368421052631576
yo
0.4485294117647059
sq
0.41972920696324956
nl
0.4152046783625731
sv
0.4129158512720157
de
0.40357852882703776
ms
0.3135011441647597
hr
0.3119266055045872
cs
0.3087557603686636
it
0.3006993006993007
ro
0.3006993006993007
sk
0.2822966507177033
sl
0.2822966507177033
lt
0.1891891891891892
lv
0.1891891891891892
pl
0.18256130790190736
reliable
true
passes
true
1
name
"Isabella"
dialogue
"**"
detectedLang
""
heavyLang
""
scores
reliable
false
passes
true
2
name
"Hunter"
dialogue
"I’m not accusing you of anything. I’m just trying to figure out what happened to the missing people."
detectedLang
"en"
heavyLang
"en"
scores
en
0.8273113708820404
tl
0.6598639455782312
no
0.40909090909090906
fr
0.37560038424591735
da
0.375
nl
0.3725868725868726
yo
0.35451837140019865
sv
0.29805615550755943
ms
0.2903930131004367
sq
0.281767955801105
is
0.27048260381593714
ca
0.25799086757990863
de
0.2562929061784897
it
0.2545871559633027
cs
0.24330616996507565
et
0.20440636474908203
ro
0.18444165621079048
pl
0.17197452229299362
pt
0.1709183673469388
sk
0.16879795396419436
lv
0.15803108808290156
sl
0.15693904020752272
hr
0.1514360313315927
lt
0.13793103448275862
fi
0.10714285714285714
tr
0.10714285714285714
es
0.09596662030598052
az
0.09090909090909091
reliable
true
passes
true
3
name
"Isabella"
dialogue
"**"
detectedLang
""
heavyLang
""
scores
reliable
false
passes
true
4
name
"Hunter"
dialogue
"What do you mean, ‘taken’?"
detectedLang
"en"
heavyLang
"en"
scores
en
0.8216409036860879
cs
0.5844875346260388
sk
0.5833333333333334
sl
0.5689655172413793
pt
0.5098039215686274
pl
0.48805460750853247
nl
0.45054945054945056
hr
0.4444444444444444
sq
0.43820224719101125
vi
0.42528735632183906
no
0.3951612903225807
sv
0.3670886075949367
tl
0.34210526315789475
yo
0.34210526315789475
reliable
true
passes
true
5
name
"Isabella"
dialogue
"**"
detectedLang
""
heavyLang
""
scores
reliable
false
passes
true
6
name
"Hunter"
dialogue
"Where were these photos found?"
detectedLang
"en"
heavyLang
"en"
scores
en
0.8412138320395202
tl
0.5730550284629982
fr
0.44717444717444715
sq
0.44168734491315137
de
0.39839572192513373
it
0.30769230769230765
es
0.26229508196721313
lt
0.26229508196721313
pt
0.26229508196721313
hi
0.2549668874172185
ca
0.2398648648648649
lv
0.2398648648648649
nl
0.2398648648648649
reliable
true
passes
true
7
name
"Isabella"
dialogue
"**"
detectedLang
""
heavyLang
""
scores
reliable
false
passes
true
8
name
"Hunter"
dialogue
"Did you report this to the police?"
detectedLang
"en"
heavyLang
"en"
scores
en
0.8361252731245449
tl
0.7551686615886833
cs
0.6987951807228916
fr
0.6382636655948554
no
0.612736660929432
sk
0.6073298429319371
hr
0.6038732394366197
sl
0.6031746031746031
da
0.5982142857142857
ro
0.5856353591160222
sv
0.576271186440678
pl
0.5631067961165048
yo
0.5202558635394456
nl
0.5161290322580645
it
0.5119305856832972
ms
0.49776785714285715
sq
0.49324324324324326
lv
0.48394495412844035
lt
0.45913461538461536
es
0.4109947643979057
et
0.3967828418230563
ca
0.3951612903225807
de
0.3951612903225807
hu
0.3869209809264305
pt
0.3869209809264305
eu
0.29467084639498436
fi
0.27184466019417475
az
0.25
tr
0.24242424242424243
reliable
true
passes
true
9
name
"Isabella"
dialogue
"**"
detectedLang
""
heavyLang
""
scores
reliable
false
passes
true
10
name
"Hunter"
dialogue
"Shit."
detectedLang
""
heavyLang
"rn"
scores
reliable
false
passes
true
11
name
"Isabella"
dialogue
"**"
detectedLang
""
heavyLang
""
scores
reliable
false
passes
true
12
name
"Hunter"
dialogue
"**"
detectedLang
""
heavyLang
""
scores
reliable
false
passes
true
13
name
"Isabella"
dialogue
"**"
detectedLang
""
heavyLang
""
scores
reliable
false
passes
true
14
name
"Hunter"
dialogue
"I think someone’s playing a game. And I’m not letting them win."
detectedLang
"en"
heavyLang
"en"
scores
en
0.8256880733944953
tl
0.7145432692307693
nl
0.48425624321389793
it
0.4583808437856328
no
0.4405182567726737
hr
0.4010088272383354
yo
0.391025641025641
ro
0.38391699092088194
da
0.37582128777923784
es
0.37582128777923784
pt
0.375
sv
0.3716931216931217
ca
0.3658210947930574
cs
0.33286516853932585
pl
0.3271954674220963
tr
0.32142857142857145
de
0.31948424068767906
hu
0.31948424068767906
is
0.309593023255814
sq
0.302496328928047
sk
0.2983751846381093
fr
0.28785607196401797
sl
0.27035330261136714
ms
0.2448330683624801
lt
0.23387096774193547
az
0.21875
lv
0.21875
vi
0.21875
et
0.2096505823627288
be
0.12844036697247707
eu
0.12844036697247707
fi
0.1171003717472119
uk
0.1171003717472119
reliable
true
passes
true
15
name
"Hunter"
dialogue
"If you see a man in a dark coat with a camera…"
detectedLang
"en"
heavyLang
"en"
scores
en
0.8032904148783977
tl
0.6405228758169935
sv
0.6352785145888594
ro
0.6148459383753502
it
0.6099290780141844
de
0.5937961595273265
nl
0.5937961595273265
ms
0.5115452930728241
sl
0.4701348747591522
ca
0.46393762183235865
et
0.4586614173228346
es
0.45652173913043476
cs
0.43298969072164945
yo
0.4294605809128631
da
0.4222689075630252
fr
0.4123931623931624
pl
0.4073275862068965
lv
0.39293598233995586
lt
0.3902439024390244
no
0.3735763097949886
hu
0.36194895591647336
az
0.358974358974359
pt
0.3373493975903614
sk
0.32098765432098764
hr
0.289405684754522
tr
0.21875
is
0.21428571428571427
sq
0.207492795389049
eu
0.20289855072463767
vi
0.20289855072463767
fi
0.1863905325443787
reliable
true
passes
true
16
name
"Hunter"
dialogue
"What the hell is she talking about?"
detectedLang
"en"
heavyLang
"en"
scores
en
0.8425472358292512
nl
0.6052631578947368
yo
0.5022123893805309
tl
0.4886363636363637
no
0.44581280788177335
is
0.4201030927835051
hu
0.4063324538258575
fr
0.39839572192513373
sv
0.39839572192513373
de
0.26229508196721313
da
0.25742574257425743
ms
0.25742574257425743
sq
0.25742574257425743
pt
0.2398648648648649
reliable
true
passes
true