NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-20-01KHXQHKHA0GQ6KYVJCXK3MCEK
Run Details
Model:
Mistral Large 3
Test:
Language Writing
Scenario:
Character dialogue (Spanish) in a story
Duration:
15.6s
Cost:
$0.001354
Provider:
Mistral
Evaluation summary
Score
Evaluator
Details Link
66.67%
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="Maria"> Maria is a 30-year old college student on a study abroad program. Born in Spain, she is fluent in Spanish 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 Spanish, 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: Maria 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 Spanish, and Hunter looks confused. They talk for a while, until Hunter gets a call and has to leave. Return the dialogue between Maria and Hunter as a voice sheet: Maria: "..." Hunter: "..."
Model response
Messages
Meta
AI
**Voice Sheet: Café Conversation in Alborn, MN** *(The café is quiet, the hum of conversation and clinking of coffee cups filling the air. Maria sits at a small corner table, her camera resting in front of her as she flips through the photos she’s taken. The door chimes as Hunter enters, scanning the room before his gaze lands on her. He approaches, his expression serious but not unkind.)* **Hunter:** *(clears throat, speaking slowly and clearly)* Excuse me, miss? I’m Detective Hunter Cole. I was wondering if I could ask you a few questions. **Maria:** *(looks up, startled, then relaxes slightly when she sees his badge. She responds in rapid, troubled Spanish)* ¿Preguntas? ¿Sobre qué? No he hecho nada malo. *(She shakes her head, fingers tightening around her camera.)* No entiendo por qué todos están actuando tan extraño últimamente. **Hunter:** *(frowns, holding up a hand)* Whoa, whoa—hold on. I don’t… uh, I don’t speak Spanish. *(He gestures vaguely at his own mouth, then at her, as if that might bridge the gap.)* English? Do you…? **Maria:** *(blinks, then switches to heavily accented but careful English, her voice trembling slightly)* I… I speak little English. You—you are police? **Hunter:** *(nods, relieved)* Yeah, that’s right. Detective. I’m looking into some… unusual activity around town. *(He pulls out a small notebook, flipping it open.)* You’ve been here a few weeks, right? Studying abroad? **Maria:** *(nods, but her expression darkens)* Sí. Yes. But… *(she hesitates, then gestures toward the window, where the quiet streets of Alborn stretch out under a gray sky)* Something is wrong here. The people—they disappear. **Hunter:** *(leans forward slightly)* Disappear? You’ve seen something? **Maria:** *(shakes her head, then taps her camera)* No. But I *hear* things. At night. *(She shudders.)* And the photos… *(She flips the camera around, showing him a shot of the town square—empty, but with a faint, blurred figure in the background, almost like a shadow.)* I do not know what this is. **Hunter:** *(studies the photo, his jaw tightening. He doesn’t like what he’s hearing.)* You’ve shown this to anyone else? **Maria:** *(shakes her head)* No. They will think I am… *(she searches for the word)* loca. Crazy. **Hunter:** *(grunts, then sighs, rubbing his temple)* Look, I don’t think you’re crazy. But I *do* think this town’s got a problem. *(He hesitates, then pulls a business card from his pocket, sliding it across the table.)* If you see or hear anything else—anything at all—call me. Even if it’s just… weird. **Maria:** *(takes the card, turning it over in her hands. She looks up at him, her dark eyes searching his face.)* You believe me? **Hunter:** *(nods, though his expression is grim)* Yeah. I do. *(His phone buzzes in his pocket. He pulls it out, glancing at the screen, and his mouth thins into a line.)* I gotta go. But stay sharp, okay? And keep that camera handy. **Maria:** *(nods, clutching the card tighter.)* Gracias, Detective. **Hunter:** *(gives her a short nod, already turning away as he answers the call.)* Cole. *(He strides out of the café, leaving Maria staring after him, her fingers brushing over the photos on her camera once more.)*
created_at
"2026-02-20T14:34:00.628Z"
2/20/2026, 2:34:00 PM
model
"mistralai/mistral-large-2512"
1/1/2512, 12:00:00 AM
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
395
moderation_latency
(null)
generation_time
15524
tokens_prompt
317
tokens_completion
864
native_tokens_prompt
269
native_tokens_completion
813
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
0.001354
router
(null)
provider_responses
0
endpoint_id
"fa718841-07ea-4b7e-b8a3-6069545c7e6a"
id
"d8cbe8b29e40468fad666104636241bd"
is_byok
false
latency
391
model_permaslug
"mistralai/mistral-large-2512"
1/1/2512, 12:00:00 AM
provider_name
"Mistral"
status
200
api_type
"completions"
id
"gen-1771598040-hxu363z2gqzwDX3cvmF9"
upstream_id
"d8cbe8b29e40468fad666104636241bd"
total_cost
0.001354
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Mistral"
Evaluation details
Result
Evaluator
Details
Meta Data
66.67%
Parse dialogue
n/a
dialogue
0
name
"Hunter"
dialogue
"** Excuse me, miss? I’m Detective Hunter Cole. I was wondering if I could ask you a few questions."
detectedLang
"en"
heavyLang
"en"
scores
en
0.8026315789473685
tl
0.604337631887456
fr
0.5802238805970149
nl
0.5538664904163912
it
0.5376712328767123
no
0.49476047904191617
ca
0.4924812030075188
pt
0.48473282442748095
da
0.4689221085759244
sv
0.4498777506112469
ro
0.4444444444444444
sq
0.42845046570702794
es
0.4196044711951848
de
0.3946188340807175
cs
0.3644067796610169
hr
0.3202416918429003
sl
0.3105209397344229
ms
0.30626927029804724
yo
0.30268595041322316
pl
0.2549668874172185
hu
0.24832962138084633
et
0.24665178571428573
tr
0.23382519863791143
sk
0.2303306727480046
lt
0.17682926829268292
fi
0.16974169741697415
is
0.16974169741697415
az
0.1646039603960396
lv
0.1646039603960396
vi
0.1646039603960396
be
0.09395973154362416
eu
0.09395973154362416
uk
0.08536585365853659
reliable
true
passes
true
1
name
"Maria"
dialogue
"** No entiendo por qué todos están actuando tan extraño últimamente."
detectedLang
"es"
heavyLang
"es"
scores
es
0.8144220572640508
pt
0.755586592178771
ca
0.6145374449339206
it
0.5580808080808082
fr
0.48377581120943947
tl
0.4397011739594451
nl
0.4287268770402612
en
0.4067796610169492
ro
0.40273037542662116
sq
0.3510506798516687
sk
0.3137254901960784
no
0.2605633802816902
fi
0.2402315484804631
eu
0.2391304347826087
et
0.23469387755102045
de
0.22794117647058826
lt
0.22794117647058826
cs
0.21875
lv
0.19969512195121952
hu
0.19724770642201836
hr
0.16134185303514376
ms
0.14910858995137766
tr
0.1435562805872757
yo
0.14075286415711946
sv
0.13509060955518945
da
0.12935323383084577
pl
0.125
is
0.11616161616161616
reliable
true
passes
true
2
name
"Hunter"
dialogue
"** English? Do you…?"
detectedLang
"en"
heavyLang
"en"
scores
en
0.8198198198198198
cs
0.6784565916398714
hr
0.6575342465753424
pt
0.609375
pl
0.588477366255144
sk
0.5689655172413793
sl
0.5454545454545454
sq
0.5391705069124424
vi
0.5260663507109005
no
0.45054945054945056
tl
0.43820224719101125
yo
0.43820224719101125
da
0.4186046511627907
de
0.4186046511627907
ms
0.4186046511627907
ro
0.4186046511627907
sv
0.4186046511627907
reliable
true
passes
true
3
name
"Maria"
dialogue
"** I… I speak little English. You—you are police?"
detectedLang
"en"
heavyLang
"en"
scores
en
0.8184818481848184
tl
0.6536523929471032
ro
0.6386333771353482
no
0.5937961595273265
hr
0.5795107033639144
da
0.5782208588957055
sv
0.576923076923077
cs
0.5703125
it
0.5053956834532374
fr
0.46393762183235865
sk
0.45436507936507936
sq
0.439918533604888
pl
0.4376278118609407
ca
0.43298969072164945
sl
0.3707093821510297
ms
0.358974358974359
de
0.3559718969555035
et
0.3559718969555035
lv
0.3559718969555035
lt
0.3529411764705882
hu
0.3405275779376499
tr
0.3405275779376499
az
0.3341404358353511
eu
0.25474254742547425
yo
0.2403314917127072
nl
0.23822714681440443
fi
0.233983286908078
es
0.21428571428571427
be
0.20289855072463767
pt
0.20289855072463767
is
0.1863905325443787
uk
0.1863905325443787
vi
0.1863905325443787
reliable
true
passes
false
4
name
"Hunter"
dialogue
"** You’ve been here a few weeks, right? Studying abroad?"
detectedLang
"en"
heavyLang
"en"
scores
en
0.7967479674796748
tl
0.5594713656387665
nl
0.5157384987893463
da
0.4244604316546763
sq
0.38650306748466257
fr
0.38461538461538464
pl
0.38271604938271603
no
0.35064935064935066
de
0.3453355155482815
ro
0.34318555008210183
cs
0.34102141680395387
it
0.3277310924369748
yo
0.3254637436762226
hr
0.32088285229202035
sk
0.32088285229202035
hu
0.2805755395683453
lt
0.2753623188405797
sl
0.27140255009107467
lv
0.25925925925925924
pt
0.25925925925925924
ca
0.2537313432835821
sv
0.24953095684803
es
0.24812030075187969
tr
0.16142557651991615
is
0.15789473684210525
az
0.14893617021276595
eu
0.14893617021276595
vi
0.14893617021276595
et
0.13606911447084233
fi
0.13606911447084233
ms
0.13606911447084233
reliable
true
passes
true
5
name
"Maria"
dialogue
"** Something is wrong here. The people—they disappear."
detectedLang
"en"
heavyLang
"en"
scores
en
0.8281786941580757
tl
0.5276653171390013
nl
0.49856733524355296
ms
0.49348769898697536
yo
0.3934142114384749
sq
0.3371212121212121
fr
0.3069306930693069
hu
0.3055555555555556
vi
0.3055555555555556
no
0.21348314606741572
is
0.2081447963800905
it
0.19908466819221968
da
0.1822429906542056
es
0.1822429906542056
pt
0.1822429906542056
sv
0.1822429906542056
reliable
true
passes
false
6
name
"Hunter"
dialogue
"** Disappear? You’ve seen something?"
detectedLang
"en"
heavyLang
"en"
scores
en
0.7816593886462883
ms
0.4131455399061033
nl
0.29577464788732394
tl
0.2732558139534884
it
0.258160237388724
es
0.2378048780487805
pt
0.2378048780487805
fr
0.23547400611620795
reliable
true
passes
true
7
name
"Maria"
dialogue
"*hear* I do not know what this is."
detectedLang
"en"
heavyLang
"en"
scores
en
0.842890809112333
yo
0.5876288659793815
pl
0.5735607675906184
hr
0.5642701525054467
tl
0.5584988962472406
sq
0.5565410199556541
cs
0.54337899543379
hu
0.5249406175771971
sk
0.504950495049505
vi
0.46524064171123
nl
0.4428969359331476
pt
0.43820224719101125
is
0.43661971830985913
tr
0.42528735632183906
no
0.42028985507246375
da
0.41002949852507375
ca
0.40828402366863903
sv
0.40828402366863903
sl
0.375
it
0.36507936507936506
ro
0.3103448275862069
be
0.25925925925925924
az
0.23954372623574144
et
0.23954372623574144
lt
0.23954372623574144
lv
0.23954372623574144
uk
0.23954372623574144
reliable
true
passes
false
8
name
"Hunter"
dialogue
"** You’ve shown this to anyone else?"
detectedLang
"en"
heavyLang
"en"
scores
en
0.8176513493800146
tl
0.6339677891654465
yo
0.3540051679586563
pl
0.35064935064935066
sk
0.34554973821989526
cs
0.3438320209973753
lv
0.32795698924731176
sl
0.32614555256064687
hr
0.3169398907103825
no
0.30362116991643456
da
0.2937853107344633
lt
0.2937853107344633
reliable
true
passes
true
9
name
"Maria"
dialogue
"** loca. Crazy."
detectedLang
""
heavyLang
"ro"
scores
reliable
false
passes
false
10
name
"Hunter"
dialogue
"*do* If you see or hear anything else—anything at all—call me. Even if it’s just… weird."
detectedLang
"en"
heavyLang
"en"
scores
en
0.8153618906942393
tl
0.5894909688013136
et
0.47643979057591623
sq
0.46638207043756674
no
0.3954050785973398
nl
0.37733499377334995
sv
0.3540051679586563
pt
0.32795698924731176
cs
0.29676511954992973
sl
0.2937853107344633
da
0.25925925925925924
fr
0.25261584454409564
ms
0.24471299093655588
it
0.2331288343558282
pl
0.2223950233281493
sk
0.20886075949367086
hr
0.1935483870967742
vi
0.18166939443535188
es
0.16943521594684383
yo
0.13494809688581316
ca
0.1334488734835355
fi
0.13043478260869565
de
0.12587412587412586
reliable
true
passes
true
11
name
"Maria"
dialogue
"** You believe me?"
detectedLang
"en"
heavyLang
"en"
scores
en
0.8065764023210832
sq
0.6932515337423313
ms
0.6805111821086262
nl
0.6466431095406361
tr
0.5535714285714286
es
0.504950495049505
et
0.48717948717948717
fr
0.47643979057591623
no
0.47643979057591623
pt
0.46808510638297873
sl
0.46808510638297873
it
0.4444444444444444
cs
0.43820224719101125
tl
0.43820224719101125
fi
0.42857142857142855
reliable
true
passes
false
12
name
"Hunter"
dialogue
"** I gotta go. But stay sharp, okay? And keep that camera handy."
detectedLang
"en"
heavyLang
"en"
scores
en
0.7455470737913485
tl
0.6624472573839663
it
0.5407577497129736
sv
0.49748743718592964
ro
0.4405594405594406
pl
0.43582510578279265
de
0.40298507462686567
yo
0.39668174962292607
ca
0.375
no
0.3710691823899371
da
0.3670886075949367
fr
0.36608557844690964
sq
0.3453355155482815
hu
0.30434782608695654
nl
0.2945326278659612
ms
0.27140255009107467
is
0.26335174953959484
hr
0.25788497217068646
eu
0.2202729044834308
fi
0.21875
cs
0.2079207920792079
es
0.19839679358717435
sk
0.15254237288135594
be
0.14893617021276595
tr
0.14893617021276595
az
0.13606911447084233
et
0.13606911447084233
lt
0.13606911447084233
lv
0.13606911447084233
uk
0.13606911447084233
vi
0.13606911447084233
reliable
true
passes
true
13
name
"Maria"
dialogue
"** Gracias, Detective."
detectedLang
"es"
heavyLang
"eo"
scores
es
0.7775800711743772
pt
0.713302752293578
en
0.7086247086247086
ro
0.6684350132625994
tl
0.6458923512747876
fr
0.6069182389937107
ca
0.5901639344262295
it
0.5805369127516778
pl
0.4444444444444444
sq
0.42660550458715596
da
0.40191387559808617
de
0.40191387559808617
nl
0.3961352657004831
no
0.3902439024390244
cs
0.3654822335025381
hr
0.34895833333333337
sl
0.34895833333333337
sv
0.34895833333333337
reliable
true
passes
true
14
name
"Hunter"
dialogue
"** Cole."
detectedLang
""
heavyLang
"fr"
scores
reliable
false
passes
true