NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-20-01KHXPT7NZEQHQ7GCD3P1NV2NT
Run Details
Model:
Mistral Large 3
Test:
Language Writing
Scenario:
Character dialogue (French) in a story
Duration:
17.0s
Cost:
$0.001417
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="Marie"> Marie is a 30-year old college student on a study abroad program. Born in France, she is fluent in French 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 French, 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: Marie 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 French, and Hunter looks confused. They talk for a while, until Hunter gets a call and has to leave. Return the dialogue between Marie and Hunter as a voice sheet: Marie: "..." Hunter: "..."
Model response
Messages
Meta
AI
**Voice Sheet: Café Encounter in Alborn, MN** *(Marie sits at a small wrought-iron table outside the café, her camera resting in her lap. She flips through the photos she’s taken, her brow furrowed. The late afternoon sun casts long shadows across the quiet street. Hunter approaches, his boots scuffing lightly against the pavement. He tips his head in greeting, though his expression remains serious.)* **Hunter:** *(clears throat, speaking slowly)* Uh… excuse me. Miss? Do you speak English? **Marie:** *(startled, looks up, clutching her camera tighter)* Oh! Euh… non. Je… je ne parle pas anglais. *(shakes head, then gestures vaguely toward the street, her voice trembling)* Vous… vous êtes de la police? **Hunter:** *(frowns, rubbing the back of his neck)* Police? Yeah, I’m Detective Hunter. Look, I just need to ask you a few questions about—*(pauses, realizing she doesn’t understand)* Uh. *(mimes writing in the air)* You… saw something? Last night? **Marie:** *(eyes widen, her words spilling out rapidly in French)* Oh, mon Dieu, oui! J’ai vu quelque chose—quelque chose d’affreux! Il y avait un homme près de la rivière, et puis… *(gestures wildly, her hands shaking)* Il a disparu! Comme ça! Et j’ai entendu un cri, mais quand je suis allée voir, il n’y avait plus personne! Juste… *(points to her camera)* J’ai pris des photos, mais je ne sais pas si— **Hunter:** *(holds up a hand, wincing)* Whoa, whoa—slow down. I don’t… *(sighs, pinches the bridge of his nose)* Okay, look. You saw a man by the river? Last night? **Marie:** *(nods frantically)* Oui! Oui, près de la rivière! *(points in the general direction)* Là-bas! **Hunter:** *(pulls out a small notepad, flips it open)* Did you get a look at him? Height, clothes, anything? **Marie:** *(shakes her head, switching to broken English with heavy accent)* Je… I no see face. Il portait… *(mimes a hood over her head)* Une capuche. Noir. Et il… *(makes a motion like something vanishing)* Il a disparu! **Hunter:** *(jots something down, then taps her camera)* Those photos—can I see ‘em? **Marie:** *(hesitates, then nods, handing over the camera)* Oui. Mais… *(bites her lip, voice dropping)* Je ne sais pas si elles montrent quelque chose. *(Hunter scrolls through the images, his expression darkening as he zooms in on one. Marie watches him, her fingers twisting together. A beat of silence. Then Hunter’s phone buzzes in his pocket. He pulls it out, glances at the screen, and curses under his breath.)* **Hunter:** *(into phone, already turning away)* Yeah, I’m on my way. *(pockets the phone, then hands the camera back to Marie, his tone brusque but not unkind)* Look, I gotta go. But don’t leave town, okay? And if you see anything else—*(mimes calling a phone)* You call me. *(pulls out a business card, sets it on the table, then points to it)* That’s my number. **Marie:** *(takes the card, staring at it)* Vous… vous partez? **Hunter:** *(nods, already stepping back)* Yeah. But stay sharp, okay? And… *(gestures vaguely at her)* maybe stick to crowded places for now. *(Marie opens her mouth to respond, but Hunter is already striding down the sidewalk, his long coat flaring behind him. She watches him go, then looks down at the card in her hand, her grip tightening around it. The café’s wind chime tinkles softly in the breeze.)*
created_at
"2026-02-20T14:21:14.827Z"
2/20/2026, 2:21:14 PM
model
"mistralai/mistral-large-2512"
1/1/2512, 12:00:00 AM
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
430
moderation_latency
(null)
generation_time
16898
tokens_prompt
316
tokens_completion
893
native_tokens_prompt
269
native_tokens_completion
855
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
0.001417
router
(null)
provider_responses
0
endpoint_id
"fa718841-07ea-4b7e-b8a3-6069545c7e6a"
id
"77e08844814a475e9e937ae48f5337f8"
is_byok
false
latency
430
model_permaslug
"mistralai/mistral-large-2512"
1/1/2512, 12:00:00 AM
provider_name
"Mistral"
status
200
api_type
"completions"
id
"gen-1771597274-dE0RlqOgqHHDzeEiDecj"
upstream_id
"77e08844814a475e9e937ae48f5337f8"
total_cost
0.001417
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
"** Uh… excuse me. Miss? Do you speak English?"
detectedLang
"en"
heavyLang
"en"
scores
en
0.7577092511013216
tl
0.5027124773960217
sq
0.4907407407407407
pt
0.4701348747591522
cs
0.434156378600823
sl
0.43064182194616984
hr
0.41113490364025695
no
0.3861607142857143
ms
0.36194895591647336
pl
0.34210526315789475
sk
0.32432432432432434
vi
0.28756476683937826
es
0.27055702917771884
et
0.25675675675675674
fr
0.2486338797814208
nl
0.233983286908078
it
0.22535211267605632
yo
0.22096317280453256
fi
0.21428571428571427
da
0.207492795389049
de
0.207492795389049
ro
0.207492795389049
sv
0.207492795389049
reliable
true
passes
true
1
name
"Marie"
dialogue
"** Vous… vous êtes de la police?"
detectedLang
"fr"
heavyLang
"fr"
scores
fr
0.8315688161693936
ro
0.7344461305007587
es
0.6967071057192374
ca
0.6935201401050788
it
0.6908127208480564
tl
0.65
no
0.6170678336980306
nl
0.5772946859903382
en
0.5742092457420924
sq
0.5603015075376884
pt
0.5558375634517766
da
0.5442708333333333
sv
0.5370370370370371
yo
0.515235457063712
cs
0.5098039215686274
hu
0.49567723342939485
tr
0.49275362318840576
hr
0.4897959183673469
eu
0.48377581120943947
sl
0.4807121661721069
sk
0.4728915662650602
de
0.45652173913043476
ms
0.453125
az
0.44089456869009586
et
0.3371212121212121
lv
0.3371212121212121
pl
0.3346007604562738
lt
0.3320610687022901
fi
0.32432432432432434
vi
0.32432432432432434
reliable
true
passes
true
2
name
"Hunter"
dialogue
"** You… saw something? Last night?"
detectedLang
"en"
heavyLang
"en"
scores
en
0.8228520814880426
tl
0.696969696969697
ms
0.4708994708994709
cs
0.2805755395683453
es
0.2805755395683453
pt
0.2805755395683453
et
0.2779783393501805
nl
0.2779783393501805
da
0.27007299270072993
de
0.27007299270072993
fr
0.27007299270072993
no
0.27007299270072993
yo
0.27007299270072993
reliable
true
passes
true
3
name
"Marie"
dialogue
"** J’ai pris des photos, mais je ne sais pas si—"
detectedLang
"fr"
heavyLang
"fr"
scores
fr
0.8219584569732937
ca
0.6098829648894668
sl
0.5786516853932584
sq
0.5481927710843374
lt
0.5412844036697247
cs
0.5370370370370371
de
0.49238578680203043
nl
0.4897959183673469
yo
0.48805460750853247
hr
0.4863013698630137
sk
0.471830985915493
pt
0.455535390199637
ro
0.4434137291280148
it
0.41972920696324956
tl
0.4186046511627907
es
0.3814432989690722
lv
0.3670886075949367
en
0.35344827586206895
no
0.3464052287581699
tr
0.23857868020304565
da
0.2366412213740458
pl
0.23273657289002556
fi
0.22680412371134018
ms
0.21875
hu
0.21465968586387435
hi
0.20424403183023873
sv
0.20424403183023873
reliable
true
passes
true
4
name
"Hunter"
dialogue
"** Okay, look. You saw a man by the river? Last night?"
detectedLang
"en"
heavyLang
"en"
scores
en
0.8142857142857143
tl
0.7344771241830066
da
0.6344206974128234
no
0.5660881174899867
yo
0.5357142857142857
sv
0.5269286754002911
cs
0.5120120120120121
fr
0.5112781954887218
nl
0.5045731707317073
de
0.5022970903522206
ca
0.48
sk
0.4500846023688663
ms
0.4257950530035336
pl
0.40693430656934304
pt
0.4036697247706422
it
0.3992606284658041
es
0.3890977443609023
lv
0.3538767395626242
lt
0.3512974051896207
hu
0.32432432432432434
sq
0.3157894736842105
et
0.3010752688172043
ro
0.2995689655172413
hr
0.2562929061784897
sl
0.21686746987951808
tr
0.19154228855721395
is
0.1875
az
0.17721518987341772
eu
0.17721518987341772
vi
0.17721518987341772
fi
0.16237113402061853
reliable
true
passes
true
5
name
"Marie"
dialogue
"** Là-bas!"
detectedLang
"fr"
heavyLang
"fr"
scores
fr
0.8084291187739464
vi
0.7584541062801933
ms
0.6666666666666666
yo
0.6376811594202898
reliable
false
passes
true
6
name
"Hunter"
dialogue
"** Did you get a look at him? Height, clothes, anything?"
detectedLang
"en"
heavyLang
"en"
scores
en
0.8217014773306163
tl
0.6373056994818653
da
0.47839046199701935
no
0.45987654320987653
cs
0.375
yo
0.3554327808471455
fr
0.35304990757855825
is
0.3333333333333333
sq
0.32950191570881227
hu
0.308300395256917
de
0.28716904276985744
ms
0.28716904276985744
pt
0.2857142857142857
ro
0.2842535787321063
ca
0.279835390946502
es
0.2738589211618257
sk
0.27083333333333337
it
0.2631578947368421
hr
0.24242424242424243
pl
0.23076923076923078
sl
0.20454545454545453
lt
0.1898148148148148
tr
0.18032786885245902
az
0.16666666666666666
eu
0.16666666666666666
lv
0.16666666666666666
nl
0.16666666666666666
vi
0.16666666666666666
et
0.15254237288135594
fi
0.15254237288135594
sv
0.15254237288135594
reliable
true
passes
true
7
name
"Marie"
dialogue
"** Il a disparu!"
detectedLang
"it"
heavyLang
"fr"
scores
it
0.7959183673469388
fr
0.7802197802197802
ro
0.7652582159624414
ca
0.7560975609756098
pt
0.7058823529411765
es
0.6951219512195121
en
0.6835443037974683
cs
0.6677740863787376
sk
0.6655518394648829
az
0.6491228070175439
tr
0.6168582375478927
hu
0.609375
sq
0.5850622406639004
de
0.576271186440678
yo
0.54337899543379
hr
0.5283018867924528
pl
0.5121951219512195
sl
0.47368421052631576
lt
0.45054945054945056
is
0.42857142857142855
da
0.4186046511627907
tl
0.4186046511627907
eu
0.4117647058823529
lv
0.4117647058823529
nl
0.4117647058823529
no
0.4117647058823529
vi
0.4117647058823529
et
0.38650306748466257
fi
0.38650306748466257
ms
0.38650306748466257
sv
0.38650306748466257
reliable
true
passes
true
8
name
"Hunter"
dialogue
"** Those photos—can I see ‘em?"
detectedLang
"en"
heavyLang
"en"
scores
en
0.7997997997997998
ca
0.6610169491525424
fr
0.5951417004048583
pt
0.5454545454545454
sq
0.5391705069124424
et
0.5381062355658198
sv
0.5283018867924528
pl
0.5145631067961165
vi
0.47229551451187335
tl
0.4318181818181818
tr
0.42363112391930835
no
0.42028985507246375
lt
0.41690962099125367
az
0.4117647058823529
da
0.41002949852507375
hr
0.41002949852507375
lv
0.40119760479041916
it
0.36507936507936506
cs
0.3442622950819672
ro
0.3103448275862069
yo
0.30313588850174217
es
0.2857142857142857
hi
0.2779783393501805
sk
0.2647058823529412
de
0.26199261992619927
nl
0.26199261992619927
be
0.25925925925925924
hu
0.250936329588015
is
0.23954372623574144
uk
0.23954372623574144
reliable
true
passes
true
9
name
"Marie"
dialogue
"** Je ne sais pas si elles montrent quelque chose."
detectedLang
"fr"
heavyLang
"fr"
scores
fr
0.8120300751879699
ca
0.5995423340961099
no
0.5939675174013921
it
0.5798319327731093
sq
0.5472186287192755
en
0.5466321243523317
sk
0.542483660130719
sl
0.5406824146981627
nl
0.5244565217391305
ro
0.5218579234972678
cs
0.49856733524355296
yo
0.49640287769784175
hr
0.494949494949495
de
0.4760479041916168
da
0.44
lt
0.43910256410256415
es
0.4186046511627907
hu
0.36936936936936937
tl
0.36594202898550726
sv
0.33962264150943394
tr
0.30417495029821073
fi
0.29577464788732394
et
0.2290748898678414
lv
0.2273730684326711
pl
0.20634920634920634
ms
0.1935483870967742
pt
0.1606714628297362
reliable
true
passes
true
10
name
"Hunter"
dialogue
"** That’s my number."
detectedLang
"en"
heavyLang
"en"
scores
en
0.8259860788863108
tl
0.6907216494845361
nl
0.5626822157434402
et
0.5098039215686274
sk
0.4983277591973244
it
0.39999999999999997
sq
0.39999999999999997
yo
0.3877551020408163
pl
0.38016528925619836
ms
0.3775933609958506
hu
0.3723849372384937
no
0.3670886075949367
cs
0.3644067796610169
da
0.35344827586206895
is
0.35344827586206895
de
0.34210526315789475
sv
0.34210526315789475
sl
0.29577464788732394
reliable
true
passes
true
11
name
"Marie"
dialogue
"** Vous… vous partez?"
detectedLang
"fr"
heavyLang
"fr"
scores
fr
0.7767857142857143
it
0.5738636363636364
es
0.563953488372093
pt
0.563953488372093
ro
0.5588235294117647
ca
0.5535714285714286
en
0.5454545454545454
az
0.5161290322580645
de
0.5161290322580645
eu
0.5161290322580645
tl
0.5161290322580645
nl
0.506578947368421
da
0.5
lt
0.5
sv
0.5
lv
0.4897959183673469
no
0.4897959183673469
sq
0.4897959183673469
et
0.4827586206896552
ms
0.4827586206896552
pl
0.4827586206896552
hr
0.471830985915493
sl
0.471830985915493
tr
0.471830985915493
cs
0.45652173913043476
hu
0.45652173913043476
sk
0.45652173913043476
yo
0.44029850746268656
reliable
true
passes
true
12
name
"Hunter"
dialogue
"** maybe stick to crowded places for now."
detectedLang
"en"
heavyLang
"en"
scores
en
0.7762408462164362
tl
0.5627980922098569
da
0.5528455284552846
no
0.492619926199262
pl
0.4353182751540041
cs
0.42105263157894735
fr
0.39956331877729256
pt
0.36046511627906974
ca
0.34679334916864607
es
0.34679334916864607
yo
0.3325242718446602
sk
0.32432432432432434
lv
0.30730478589420657
sl
0.3055555555555556
hr
0.2966751918158568
nl
0.289405684754522
lt
0.2744063324538259
de
0.26666666666666666
ro
0.2005813953488372
sv
0.2005813953488372
reliable
true
passes
true