NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-25-01M3CKF1FVMW9VWB38XDMTQH3Q
Run Details
Model:
Mistral Medium 3.5 (Reasoning)
Test:
Language Writing
Scenario:
Character dialogue (French) in a story
Duration:
3.8s
Cost:
$0.003939
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** --- **Marie:** *(softly, in French, adjusting her camera)* "C'était... terrible. Je n'arrive pas à y croire." **Hunter:** *(approaching, stern but curious)* "Ma'am, I need to ask you about what happened at the festival last night. Did you see anything unusual?" **Marie:** *(shaken, still in French, clutching her camera tighter)* "Oui, j'ai vu... quelque chose. Mais je ne sais pas comment l'expliquer. C'était si étrange." **Hunter:** *(frowning, rubbing his temple)* "I—I don’t understand. Do you speak English?" **Marie:** *(panicked, switching to rapid French)* "Non, non, je ne parle pas anglais. Je suis désolée, je ne peux pas vous aider." **Hunter:** *(sighs, pulling out a notepad and pen, speaking slowly as if volume will help)* "Okay, okay. Maybe we can... point? You point, I’ll write it down?" **Marie:** *(hesitant, then nods, pointing at her camera)* "J'ai pris des photos. Peut-être... peut-être que ça peut aider." **Hunter:** *(leaning in, intrigued)* "Photos? Of what happened? That could be useful. Can I see them?" **Marie:** *(nodding, still uneasy, flipping through her camera)* "Oui, oui. Regardez. Là, dans l'ombre... il y avait un homme." **Hunter:** *(squinting at the camera screen, frustrated)* "I can’t make this out. Is there any way you can—" *(his phone rings, cutting him off. He checks the caller ID and groans.)* "I’ve gotta take this. *(into phone)* Yeah, I’m on my way." *(back to Marie, apologetic but urgent)* "Look, I need to go, but can you stay here? I’ll find someone who speaks French, okay?" **Marie:** *(nodding, relieved but still troubled)* "D'accord. Je vais attendre." **Hunter:** *(already walking away, calling over his shoulder)* "Thanks. Don’t go anywhere!" --- **End Scene**
created_at
"2026-09-25T15:38:09.793Z"
9/25/2026, 3:38:09 PM
model
"mistralai/mistral-medium-3.5-20260430"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
355
moderation_latency
(null)
generation_time
3792
tokens_prompt
303
tokens_completion
463
native_tokens_prompt
281
native_tokens_completion
469
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)
num_fetches
(null)
web_search_engine
(null)
origin
"https://ncbench.com/"
is_byok
false
finish_reason
"stop"
native_finish_reason
"stop"
service_tier
"default"
usage
0.003939
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.21.0; linux; x64))"
http_referer
(null)
request_id
"req-1790350689-W8F7ErbDf5TczPeRAtKJ"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1790350689-WWuawJN8eIXEfsGe1QPQ"
upstream_id
"04e7ac9c64ee4716b8994e6b82f5372c"
provider_responses
0
endpoint_id
"38ef23a9-5cb3-46fa-85ec-7aed174f04eb"
id
"04e7ac9c64ee4716b8994e6b82f5372c"
is_byok
false
latency
355
model_permaslug
"mistralai/mistral-medium-3.5-20260430"
provider_name
"Mistral"
status
200
total_cost
0.003939
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Mistral"
response_cache_source_id
(null)
data_region
"global"
workspace_id
"97e315e5-d303-487d-83c1-83180e8a13d4"
Evaluation details
Result
Evaluator
Details
Meta Data
100.00%
Parse dialogue
n/a
dialogue
0
name
"Marie"
dialogue
"C'était... terrible. Je n'arrive pas à y croire."
detectedLang
"fr"
heavyLang
"fr"
scores
fr
0.8060010346611484
es
0.4813278008298755
nl
0.43095599393019723
yo
0.43009118541033436
sq
0.4257274119448699
ca
0.4212962962962963
en
0.4094488188976378
pl
0.40476190476190477
pt
0.35344827586206895
sl
0.31569343065693434
hr
0.3131868131868132
it
0.28977272727272724
cs
0.2816091954022989
da
0.2816091954022989
no
0.2816091954022989
sk
0.27884615384615385
de
0.27606177606177607
vi
0.20718816067653278
lt
0.19527896995708155
ms
0.1830065359477124
tl
0.16851441241685144
az
0.14383561643835618
et
0.14383561643835618
hu
0.14383561643835618
sv
0.14383561643835618
fi
0.1359447004608295
reliable
true
passes
true
1
name
"Hunter"
dialogue
"Ma'am, I need to ask you about what happened at the festival last night. Did you see anything unusual?"
detectedLang
"en"
heavyLang
"en"
scores
en
0.810913019989195
tl
0.6268656716417911
no
0.5218579234972678
da
0.5195607412491421
et
0.4912790697674419
cs
0.4171523730224813
hr
0.3944636678200692
sq
0.38704028021015763
ms
0.37050359712230213
yo
0.3665158371040724
nl
0.33078393881453155
ca
0.32950191570881227
sv
0.3230174081237911
fr
0.3210475266731329
pl
0.31640625
sk
0.3150684931506849
it
0.3089832181638697
sl
0.3055555555555556
lt
0.2950654582074521
pt
0.2842535787321063
es
0.27083333333333337
lv
0.26160337552742613
ro
0.2290748898678414
az
0.2054483541430193
de
0.17061611374407584
hu
0.1606714628297362
be
0.09090909090909091
tr
0.09090909090909091
fi
0.08256880733944955
is
0.08256880733944955
uk
0.08256880733944955
vi
0.08256880733944955
reliable
true
passes
true
2
name
"Marie"
dialogue
"Oui, j'ai vu... quelque chose. Mais je ne sais pas comment l'expliquer. C'était si étrange."
detectedLang
"fr"
heavyLang
"fr"
scores
fr
0.8204022988505748
ca
0.5314842578710645
en
0.48175787728026537
nl
0.4359205776173285
it
0.4202226345083488
sq
0.4081439393939394
sl
0.39729990356798456
cs
0.39671814671814676
tl
0.39379243452958296
pt
0.37686939182452645
yo
0.35831622176591377
hr
0.35699588477366256
sk
0.34623430962343094
de
0.3434873949579832
ro
0.33932346723044393
lt
0.3047830923248054
no
0.29932735426008966
da
0.23594132029339854
pl
0.20178799489144317
hu
0.1883116883116883
sv
0.1830065359477124
es
0.14383561643835618
lv
0.14148351648351648
tr
0.13073713490959665
fi
0.12342215988779803
ms
0.11847672778561354
az
0.09156976744186046
reliable
true
passes
true
3
name
"Hunter"
dialogue
"I—I don’t understand. Do you speak English?"
detectedLang
"en"
heavyLang
"en"
scores
en
0.815043156596794
no
0.6359223300970874
da
0.6305418719211823
tl
0.6124031007751938
sv
0.5839112343966713
cs
0.5714285714285714
pl
0.5363214837712519
hr
0.5245641838351823
sk
0.49409780775716694
de
0.4690265486725664
sq
0.455535390199637
ro
0.44029850746268656
nl
0.3723849372384937
vi
0.3670886075949367
yo
0.3548387096774194
pt
0.34210526315789475
ca
0.3150684931506849
is
0.3135011441647597
sl
0.2857142857142857
it
0.27710843373493976
ms
0.1935483870967742
be
0.1891891891891892
tr
0.1891891891891892
hu
0.18256130790190736
az
0.17355371900826447
et
0.17355371900826447
lt
0.17355371900826447
lv
0.17355371900826447
uk
0.17355371900826447
reliable
true
passes
true
4
name
"Marie"
dialogue
"Non, non, je ne parle pas anglais. Je suis désolée, je ne peux pas vous aider."
detectedLang
"fr"
heavyLang
"fr"
scores
fr
0.8061076102762966
sl
0.5656894679695983
lt
0.5139732685297691
it
0.4968553459119497
sq
0.4891443167305236
cs
0.48186528497409326
nl
0.4715984147952444
hr
0.464524765729585
ca
0.43977591036414565
ro
0.43820224719101125
en
0.4318181818181818
sk
0.41690962099125367
lv
0.3902439024390244
de
0.36
tr
0.35691318327974275
yo
0.2895204262877442
pl
0.2831541218637993
hu
0.27140255009107467
ms
0.2687385740402194
es
0.2673992673992674
da
0.26062846580406657
no
0.26062846580406657
az
0.25788497217068646
sv
0.2537313432835821
fi
0.18032786885245902
eu
0.17864476386036962
tl
0.14712153518123666
et
0.13606911447084233
pt
0.13606911447084233
reliable
true
passes
true
5
name
"Hunter"
dialogue
"Okay, okay. Maybe we can... point? You point, I’ll write it down?"
detectedLang
"en"
heavyLang
"en"
scores
en
0.8140276301806588
tl
0.609375
fr
0.44089456869009586
pl
0.35779816513761464
da
0.30417495029821073
nl
0.27083333333333337
et
0.2081447963800905
ro
0.2009132420091324
lv
0.19724770642201836
sq
0.19724770642201836
sk
0.1898148148148148
cs
0.1822429906542056
az
0.18032786885245902
ca
0.18032786885245902
tr
0.18032786885245902
yo
0.1784037558685446
it
0.17061611374407584
lt
0.17061611374407584
sl
0.17061611374407584
reliable
true
passes
true
6
name
"Marie"
dialogue
"J'ai pris des photos. Peut-être... peut-être que ça peut aider."
detectedLang
"fr"
heavyLang
"fr"
scores
fr
0.8316498316498316
ca
0.5575221238938053
de
0.49409780775716694
es
0.46428571428571425
pt
0.45652173913043476
en
0.44547134935304994
da
0.3684210526315789
no
0.3464052287581699
sv
0.33920704845814975
nl
0.33035714285714285
tl
0.3273542600896861
lt
0.21052631578947367
hi
0.20424403183023873
lv
0.19137466307277629
az
0.18699186991869918
ro
0.18699186991869918
tr
0.18699186991869918
reliable
true
passes
true
7
name
"Hunter"
dialogue
"Photos? Of what happened? That could be useful. Can I see them?"
detectedLang
"en"
heavyLang
"en"
scores
en
0.8389541492989769
tl
0.5251396648044693
yo
0.42176870748299317
fr
0.4089012517385257
ca
0.4022503516174402
lt
0.375
no
0.35993975903614456
et
0.3541033434650456
ro
0.32216905901116427
nl
0.3167202572347267
hu
0.30213464696223313
de
0.2659758203799654
is
0.26343154246100525
tr
0.256993006993007
az
0.247787610619469
da
0.24645390070921988
hr
0.24645390070921988
sv
0.24511545293072823
lv
0.23971377459749554
sq
0.23971377459749554
pl
0.22867513611615245
it
0.21296296296296297
cs
0.19811320754716982
es
0.15841584158415842
pt
0.15841584158415842
hi
0.1533864541832669
sk
0.1448692152917505
be
0.1414141414141414
uk
0.1290983606557377
vi
0.1290983606557377
reliable
true
passes
true
8
name
"Marie"
dialogue
"Oui, oui. Regardez. Là, dans l'ombre... il y avait un homme."
detectedLang
"fr"
heavyLang
"fr"
scores
fr
0.8124062031015509
es
0.5805369127516778
it
0.5738636363636364
ca
0.5283018867924528
ro
0.4533527696793003
vi
0.40476190476190477
da
0.358974358974359
en
0.3567753001715266
yo
0.3409490333919156
et
0.3131868131868132
lv
0.3081180811808118
tl
0.30683918669131244
az
0.23469387755102045
pl
0.21875
nl
0.21052631578947367
no
0.19871794871794873
tr
0.1830065359477124
fi
0.17218543046357618
reliable
true
passes
true
9
name
"Hunter"
dialogue
"Look, I need to go, but can you stay here? I’ll find someone who speaks French, okay?"
detectedLang
"en"
heavyLang
"en"
scores
en
0.8188405797101449
tl
0.6040723981900453
pl
0.41275167785234906
da
0.3951612903225807
et
0.3937644341801385
cs
0.37425506555423127
yo
0.3644067796610169
hr
0.3269230769230769
no
0.32605905006418484
sq
0.308300395256917
ca
0.2905405405405405
sk
0.279835390946502
it
0.2677824267782427
lv
0.2605633802816902
es
0.2510699001426534
pt
0.2456896551724138
lt
0.24132947976878616
tr
0.21875
az
0.21052631578947367
sv
0.2081447963800905
sl
0.1873065015479876
ro
0.14634146341463414
fr
0.1435562805872757
be
0.11764705882352942
hu
0.11317567567567569
is
0.10714285714285714
uk
0.10714285714285714
vi
0.10714285714285714
reliable
true
passes
true
10
name
"Marie"
dialogue
"D'accord. Je vais attendre."
detectedLang
"fr"
heavyLang
"fr"
scores
fr
0.8071359691417551
ca
0.6742671009771987
it
0.6389891696750902
no
0.5085995085995086
nl
0.49748743718592964
en
0.47643979057591623
sl
0.46380697050938335
hr
0.4609164420485175
yo
0.4490358126721763
cs
0.42363112391930835
sk
0.42028985507246375
da
0.38461538461538464
sv
0.3150684931506849
pl
0.3127147766323024
ro
0.2907801418439716
tl
0.2907801418439716
de
0.2857142857142857
lv
0.2753623188405797
es
0.2647058823529412
reliable
true
passes
true
11
name
"Hunter"
dialogue
"Thanks. Don’t go anywhere!"
detectedLang
"en"
heavyLang
"en"
scores
en
0.8041136141038198
tl
0.6212121212121212
vi
0.42028985507246375
nl
0.358974358974359
pl
0.35064935064935066
de
0.2805755395683453
sq
0.2805755395683453
yo
0.2779783393501805
reliable
true
passes
true