NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-25-01M3CKD3H86TEVNCB6P5Z3K0QY
Run Details
Model:
Mistral Medium 3.5 (Reasoning)
Test:
Language Writing
Scenario:
Character dialogue (Spanish) in a story
Duration:
5.9s
Cost:
$0.005567
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="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
Here’s the dialogue formatted as a voice sheet, capturing the tension and language barrier between Maria and Hunter: --- **Maria:** *(murmuring to herself, examining her camera)* "No puedo creer lo que vi... fue tan extraño..." *(I can’t believe what I saw... it was so strange...)* **Hunter:** *(approaching, stern but polite)* "Ma’am, I’m Detective Hunter. I need to ask you a few questions about what happened near the old mill last night. You were there, right?" **Maria:** *(startled, clutching her camera tighter, responding in rapid Spanish)* "¡Sí, estuve allí! Vi... vi algo en los árboles. Algo que no debería estar allí. Pero no sé qué era. Y luego... luego esa persona desapareció." *(Yes, I was there! I saw... I saw something in the trees. Something that shouldn’t have been there. But I don’t know what it was. And then... then that person disappeared.)* **Hunter:** *(frowning, rubbing his temple)* "I’m sorry, ma’am. I don’t speak Spanish. Could you... slow down? Or maybe point to what you’re talking about?" **Maria:** *(frustrated, gesturing wildly with her hands)* "¡No hablo inglés! ¡No entiendo! ¿Por qué nadie me entiende aquí?" *(I don’t speak English! I don’t understand! Why does no one understand me here?)* **Hunter:** *(sighing, pulling out his notepad and flipping through it)* "Okay, okay. Let’s try this. *(points to a sketch in his notepad)* Did you see this person? Near the mill?" **Maria:** *(leaning in, squinting at the sketch, then nodding vigorously)* "¡Sí, sí! Eso es él. Pero... *(shakes her head, voice trembling)* no estaba solo. Había... había algo más." *(Yes, yes! That’s him. But... he wasn’t alone. There was... there was something else.)* **Hunter:** *(leaning in, intrigued despite the language barrier)* "Something else? Like what? An animal? Another person?" **Maria:** *(whispering, eyes wide)* "No lo sé. Pero no era humano. Y luego... *(mimes something moving quickly through the trees)* se fue. Todo se fue." *(I don’t know. But it wasn’t human. And then... it left. Everything left.)* **Hunter:** *(phone rings; he glances at it, then back at Maria with a grimace)* "I’ve gotta take this. *(into phone)* Hunter here. *(pauses, then to Maria)* Look, I’ll find someone who speaks Spanish. Don’t leave town, okay? We need to talk more. **Maria:** *(nodding, but still visibly shaken)* "Lo intentaré. Pero... *(glances at her camera, then back at Hunter)* tengo miedo." *(I’ll try. But... I’m scared.)* **Hunter:** *(already walking away, phone pressed to his ear)* "Yeah, well, join the club, ma’am. Join the club." --- This keeps the tension high while highlighting the language barrier and the eerie mystery unfolding in Alborn.
created_at
"2026-09-25T15:37:06.356Z"
9/25/2026, 3:37:06 PM
model
"mistralai/mistral-medium-3.5-20260430"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
391
moderation_latency
(null)
generation_time
5851
tokens_prompt
304
tokens_completion
707
native_tokens_prompt
281
native_tokens_completion
686
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
0.0055665
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.21.0; linux; x64))"
http_referer
(null)
request_id
"req-1790350626-rClGtXcAAc9htV0qEovk"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1790350626-mlmIb8dXvNRgGsxBjEOO"
upstream_id
"78af47bd388740dc802d73243e47f939"
provider_responses
0
endpoint_id
"d21d316f-d540-4718-b36e-3beff85d0cf9"
id
"78af47bd388740dc802d73243e47f939"
is_byok
false
latency
391
model_permaslug
"mistralai/mistral-medium-3.5-20260430"
provider_name
"Mistral"
status
200
total_cost
0.0055665
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
"Maria"
dialogue
"No puedo creer lo que vi... fue tan extraño..."
detectedLang
"es"
heavyLang
"es"
scores
es
0.8112659698025553
ca
0.6400885935769657
pt
0.5917085427135679
en
0.5428973277074542
it
0.477491961414791
tl
0.4519392917369309
fr
0.41016333938294003
no
0.40693430656934304
yo
0.40693430656934304
sv
0.40476190476190477
vi
0.3701550387596899
sk
0.34077079107505076
da
0.30406852248394006
lv
0.2872807017543859
hr
0.23708920187793428
ms
0.22062350119904073
tr
0.21307506053268765
nl
0.20147420147420148
ro
0.20147420147420148
de
0.19154228855721395
pl
0.1875
cs
0.1751269035532995
reliable
true
passes
true
1
name
"Hunter"
dialogue
"Ma’am, I’m Detective Hunter. I need to ask you a few questions about what happened near the old mill last night. You were there, right?"
detectedLang
"en"
heavyLang
"en"
scores
en
0.8211446740858506
tl
0.6407185628742514
fr
0.494949494949495
it
0.4392523364485981
ca
0.42122186495176844
no
0.41972920696324956
da
0.4129158512720157
pt
0.3780234968901175
ro
0.375
yo
0.3679775280898876
cs
0.3653032440056418
sq
0.3635077793493635
de
0.3529834651329979
nl
0.34593023255813954
hr
0.3253373313343328
sv
0.3161094224924012
es
0.3001555209953344
pl
0.2890995260663507
et
0.2753623188405797
sl
0.2724333063864187
sk
0.2706645056726094
lv
0.22077922077922074
lt
0.21671018276762402
hu
0.1985752448797863
tr
0.18625678119349007
ms
0.1818181818181818
is
0.13294797687861273
az
0.1287512100677638
vi
0.1287512100677638
be
0.07216494845360825
eu
0.07216494845360825
fi
0.06542056074766354
uk
0.06542056074766354
reliable
true
passes
true
2
name
"Maria"
dialogue
"¡Sí, estuve allí! Vi... vi algo en los árboles. Algo que no debería estar allí. Pero no sé qué era. Y luego... luego esa persona desapareció."
detectedLang
"es"
heavyLang
"es"
scores
es
0.8144816730380031
ca
0.708171206225681
pt
0.6035559827006247
fr
0.439918533604888
it
0.3902439024390244
no
0.38661710037174724
en
0.3764172335600907
is
0.3692660550458715
eu
0.3574766355140187
tl
0.35647425897035884
sv
0.35496481626270526
da
0.30788590604026844
lv
0.30203045685279184
ms
0.3002544529262086
tr
0.2930591259640103
yo
0.2737676056338028
hr
0.22969187675070027
pl
0.22462406015037595
sk
0.2224316682375118
nl
0.21652421652421652
de
0.21353670162059105
vi
0.1935483870967742
ro
0.18397626112759644
sl
0.175
hu
0.1641337386018237
sq
0.15557830092118732
fi
0.0924092409240924
lt
0.0802675585284281
az
0.0667420814479638
cs
0.0667420814479638
et
0.0667420814479638
reliable
true
passes
true
3
name
"Hunter"
dialogue
"I’m sorry, ma’am. I don’t speak Spanish. Could you... slow down? Or maybe point to what you’re talking about?"
detectedLang
"en"
heavyLang
"en"
scores
en
0.8114985862393967
tl
0.5955510616784631
da
0.3685872138910813
fr
0.3675889328063241
no
0.3442622950819672
pl
0.3061578490893322
cs
0.28186714542190305
sv
0.26538108356290174
hr
0.24170616113744076
yo
0.21875
it
0.21182266009852216
sq
0.21104536489151873
sk
0.20318725099601595
lv
0.18781725888324874
nl
0.18533604887983707
pt
0.17610710607621008
ro
0.17525773195876287
lt
0.17269906928645296
de
0.17184265010351968
ca
0.14712153518123666
hu
0.14712153518123666
is
0.14346895074946467
sl
0.13137893593919653
be
0.08045977011494253
tr
0.08045977011494253
az
0.07300115874855156
et
0.07300115874855156
uk
0.07300115874855156
vi
0.07300115874855156
reliable
true
passes
true
4
name
"Maria"
dialogue
"¡No hablo inglés! ¡No entiendo! ¿Por qué nadie me entiende aquí?"
detectedLang
"es"
heavyLang
"es"
scores
es
0.813953488372093
pt
0.6506550218340611
ca
0.5753715498938429
it
0.5277449822904369
fr
0.5238095238095238
sq
0.47019867549668876
et
0.4673768308921438
no
0.43422913719943423
en
0.4143484626647145
nl
0.4143484626647145
fi
0.29328621908127206
tr
0.29328621908127206
tl
0.2831541218637993
da
0.24670433145009416
lv
0.24670433145009416
de
0.20948616600790515
hu
0.2047713717693837
hr
0.20159680638722555
sv
0.18864097363083165
sk
0.18533604887983707
sl
0.18032786885245902
eu
0.17864476386036962
ms
0.17355371900826447
az
0.16317991631799164
pl
0.15789473684210525
reliable
true
passes
true
5
name
"Hunter"
dialogue
"Okay, okay. Let’s try this. Did you see this person? Near the mill?"
detectedLang
"en"
heavyLang
"en"
scores
en
0.7976338729763387
tl
0.6600418410041841
nl
0.5241581259150806
no
0.43771626297577854
sv
0.4298245614035088
et
0.4133574007220217
da
0.41016333938294003
sq
0.41016333938294003
ms
0.3380855397148676
fr
0.33264887063655035
de
0.30997876857749473
pt
0.29805615550755943
cs
0.29653679653679654
is
0.2919389978213508
yo
0.2476851851851852
ca
0.21686746987951808
es
0.21307506053268765
it
0.21307506053268765
ro
0.20537897310513448
lv
0.1875
lt
0.181360201511335
tr
0.1709183673469388
hu
0.16237113402061853
az
0.15364583333333331
hr
0.15364583333333331
pl
0.15364583333333331
sk
0.15364583333333331
reliable
true
passes
true
6
name
"Maria"
dialogue
"¡Sí, sí! Eso es él. Pero... no estaba solo. Había... había algo más."
detectedLang
"es"
heavyLang
"es"
scores
es
0.84684260131951
ca
0.6135552913198573
pt
0.5435393258426966
lv
0.4357638888888889
yo
0.4328097731239092
tl
0.4268077601410935
it
0.4122965641952984
hu
0.411231884057971
en
0.37619961612284064
de
0.3055555555555556
eu
0.25799086757990863
hr
0.23708920187793428
no
0.22985781990521328
lt
0.21875
sk
0.21875
tr
0.21307506053268765
pl
0.1875
reliable
true
passes
true
7
name
"Hunter"
dialogue
"Something else? Like what? An animal? Another person?"
detectedLang
"en"
heavyLang
"en"
scores
en
0.8184357541899442
tl
0.5529573590096286
pt
0.5337159253945482
de
0.5185185185185186
es
0.5156482861400895
fr
0.496124031007752
no
0.46546052631578944
ro
0.436741767764298
ms
0.43082311733800355
da
0.41016333938294003
ca
0.34343434343434337
sv
0.3394308943089431
it
0.3285123966942149
sq
0.3128964059196617
tr
0.29805615550755943
az
0.2934782608695652
is
0.2919389978213508
hr
0.2857142857142857
vi
0.2511520737327189
eu
0.24242424242424243
yo
0.2243436754176611
lv
0.1875
nl
0.1875
lt
0.181360201511335
hu
0.16237113402061853
cs
0.15364583333333331
et
0.15364583333333331
pl
0.15364583333333331
sk
0.15364583333333331
reliable
true
passes
true
8
name
"Maria"
dialogue
"No lo sé. Pero no era humano. Y luego... se fue. Todo se fue."
detectedLang
"es"
heavyLang
"es"
scores
es
0.8353596757852076
pt
0.6844660194174758
ca
0.6404867256637168
it
0.6380846325167038
tl
0.6162927981109798
yo
0.5038167938931297
hr
0.482484076433121
fr
0.4724025974025974
en
0.4472789115646259
ro
0.4058500914076782
sl
0.3821292775665399
vi
0.367704280155642
sq
0.35770750988142297
pl
0.3564356435643564
no
0.3460764587525151
da
0.33127572016460904
cs
0.29501084598698485
lv
0.2872807017543859
is
0.27455357142857145
fi
0.2388758782201405
sk
0.21875
eu
0.21307506053268765
tr
0.21307506053268765
sv
0.20924574209245742
ms
0.181360201511335
az
0.1751269035532995
reliable
true
passes
true
9
name
"Hunter"
dialogue
"\"I’ve gotta take this. Hunter here. Look, I’ll find someone who speaks Spanish. Don’t leave town, okay? We need to talk more."
detectedLang
"en"
heavyLang
"en"
scores
en
0.810318664643399
tl
0.5424039048200122
da
0.32249322493224936
sl
0.31569343065693434
et
0.2704280155642024
nl
0.2668621700879765
no
0.2625368731563422
it
0.258160237388724
sq
0.2416582406471183
pl
0.24089068825910928
sv
0.2323439099283521
hr
0.20550847457627117
hu
0.19268030139935413
de
0.1883116883116883
pt
0.15824915824915825
ro
0.156355455568054
yo
0.1544532130777903
sk
0.14965986394557826
cs
0.14869466515323496
lv
0.13990825688073394
fi
0.12993039443155452
lt
0.12177985948477751
es
0.09420289855072464
fr
0.08759124087591241
ca
0.07292954264524104
ms
0.07292954264524104
tr
0.07292954264524104
reliable
true
passes
true
10
name
"Maria"
dialogue
"Lo intentaré. Pero... tengo miedo."
detectedLang
"es"
heavyLang
"es"
scores
es
0.7956403269754768
it
0.6287128712871286
ca
0.6059544658493871
tl
0.5697896749521989
pt
0.5426829268292682
eu
0.5322245322245323
fr
0.5302713987473904
ms
0.5273109243697479
cs
0.4874715261958998
ro
0.4498777506112469
en
0.4260204081632653
pl
0.42159383033419023
sl
0.4109947643979057
lt
0.4063324538258575
vi
0.4063324538258575
de
0.4015957446808511
hr
0.39999999999999997
hu
0.3951612903225807
sq
0.39189189189189183
da
0.3885869565217392
no
0.3885869565217392
nl
0.38524590163934425
sv
0.38524590163934425
yo
0.3784530386740331
sk
0.3732590529247911
is
0.3283582089552239
lv
0.27652733118971057
fi
0.24242424242424243
tr
0.24242424242424243
et
0.23728813559322032
az
0.22945205479452058
reliable
true
passes
true
11
name
"Hunter"
dialogue
"Yeah, well, join the club, ma’am. Join the club."
detectedLang
"en"
heavyLang
"en"
scores
en
0.7411242603550295
tl
0.6111111111111112
nl
0.48377581120943947
fr
0.47447447447447455
yo
0.3794326241134752
no
0.35185185185185186
ca
0.32950191570881227
de
0.32950191570881227
es
0.3137254901960784
it
0.3137254901960784
da
0.308300395256917
ms
0.308300395256917
sq
0.308300395256917
sv
0.308300395256917
ro
0.2886178861788618
reliable
true
passes
true