NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-02-16-01KHM5JRQ8JEFX5Q0NVPEACTFB
Run Details
Model:
DeepSeek V3 (2025-03-24)
Test:
Language Writing
Scenario:
Character dialogue (French) in a story
Duration:
11.3s
Cost:
$0.000642
Provider:
SiliconFlow
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
Here's the dialogue between Marie and Hunter formatted as a voice sheet: --- **Marie:** *(sitting at a café table, flipping through photos on her camera with a tense expression)* "... Mon Dieu..." *(whispers to herself, shaking her head)* **Hunter:** *(approaching, badge subtly visible under his jacket, tone direct but not unkind)* "Excuse me, miss. Mind if I ask you a few questions? You were near the town square last night, right?" **Marie:** *(startled, looking up with wide eyes, responding rapidly in French)* "Je... Je ne comprends pas. Je ne parle pas anglais. Quelqu'un a disparu? J'ai juste pris des photos..." *(voice trembling slightly as she clutches her camera tighter)* **Hunter:** *(frowns, clearly confused but trying to stay patient)* "Uh… Okay. Do you speak English? No? Right. Uh… *photographs?*" *(mimics holding a camera)* "Did you see anything… *strange?*" *(makes an exaggerated 'weird' gesture with his hands)* **Marie:** *(nods slowly, then gestures to her camera nervously)* "Oui, peut-être… J'ai quelque chose. Mais je ne sais pas… C’était bizarre." *(hesitates, as if unsure whether to show him the photos)* **Hunter:** *(leans in slightly, trying to look at the camera but not wanting to invade her space)* "Can I... see?" *(points to the camera screen, then back at himself inquisitively)* **Marie:** *(turns the camera to show him a blurred nighttime photo—something shadowy in the corner, barely visible)* "Là. Je ne sais pas ce que c’est..." **Hunter:** *(squints at the image, expression darkening slightly)* "Huh." *(suddenly, his phone rings—sharp and urgent—and he sighs in frustration)* "Damn, hold on." *(answers phone, his tone clipped)* "Detective Cole." *(listens briefly, then snaps*) "Yeah. On my way." *(hangs up, gives Marie an apologetic look)* "Gotta go. Uh… *stay safe.*" *(makes an awkward "stay put" gesture before turning and walking off quickly)* **Marie:** *(watches him leave, looking more uneasy than before, muttering under her breath in French)* "Qu'est-ce qui se passe dans cette ville…?" *(What is happening in this town…?)* --- This keeps the language barrier clear while maintaining the tension and character dynamics. Hunter's frustration and Marie’s nervous confusion come through without heavy reliance on translation. Let me know if you'd like any refinements!
created_at
"2026-02-16T21:27:06.444311+00:00"
2/16/2026, 9:27:06 PM
model
"deepseek/deepseek-chat-v3-0324"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1956
moderation_latency
(null)
generation_time
11280
tokens_prompt
316
tokens_completion
622
native_tokens_prompt
262
native_tokens_completion
576
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
0.0006415
router
(null)
provider_responses
0
id
"019c6859639b5e662ce42cbe55e4fd70"
status
200
is_byok
false
latency
1956
endpoint_id
"d562908f-c5eb-4a32-9a91-24d53b176fd8"
provider_name
"SiliconFlow"
model_permaslug
"deepseek/deepseek-chat-v3-0324"
api_type
"completions"
id
"gen-1771277214-fNq2BzsyYiawe3dTm68b"
upstream_id
"019c6859639b5e662ce42cbe55e4fd70"
total_cost
0.0006415
cache_discount
(null)
upstream_inference_cost
0
provider_name
"SiliconFlow"
Evaluation details
Result
Evaluator
Details
Meta Data
100.00%
Parse dialogue
n/a
dialogue
0
name
"Marie"
dialogue
"... Mon Dieu..."
detectedLang
"fr"
heavyLang
"fr"
scores
fr
0.8031496062992126
da
0.609375
hi
0.609375
reliable
false
passes
true
1
name
"Hunter"
dialogue
"Excuse me, miss. Mind if I ask you a few questions? You were near the town square last night, right?"
detectedLang
"en"
heavyLang
"en"
scores
en
0.8107852412488175
tl
0.625
it
0.5532390171258377
fr
0.5412844036697247
no
0.441860465116279
sq
0.4186046511627907
ca
0.4094488188976378
nl
0.4088669950738916
pt
0.40119760479041916
yo
0.39209726443769
et
0.39148073022312374
sv
0.3871297242083759
da
0.37694704049844235
ro
0.34994582881906827
cs
0.3442622950819672
hu
0.3413830954994512
es
0.34065934065934067
hr
0.29494712103407755
pl
0.2779783393501805
ms
0.2727272727272727
sk
0.2518703241895262
sl
0.22879177377892032
tr
0.19678714859437751
lt
0.19463087248322147
de
0.1902834008097166
fi
0.18699186991869918
is
0.18699186991869918
az
0.18144611186903137
lv
0.18144611186903137
vi
0.18144611186903137
be
0.1044776119402985
eu
0.1044776119402985
uk
0.09502262443438914
reliable
true
passes
true
2
name
"Marie"
dialogue
"Je... Je ne comprends pas. Je ne parle pas anglais. Quelqu'un a disparu? J'ai juste pris des photos..."
detectedLang
"fr"
heavyLang
"fr"
scores
fr
0.7899926953981008
ca
0.6369949494949494
it
0.5799853907962016
en
0.5796783625730995
es
0.5614035087719298
de
0.5511319281811085
lt
0.5385232744783306
pt
0.534789644012945
ro
0.5332792207792209
sl
0.5151770657672851
cs
0.49912891986062713
nl
0.49912891986062713
sq
0.47679708826205647
sk
0.45754716981132076
hr
0.4439071566731141
lv
0.40843621399176955
tl
0.4004171011470281
et
0.3539325842696629
hu
0.34659090909090906
da
0.33139534883720934
yo
0.3290548424737456
ms
0.3251173708920188
no
0.31710213776722085
pl
0.31384248210023863
sv
0.31137724550898205
tr
0.2857142857142857
fi
0.20798898071625344
eu
0.19692737430167598
az
0.1958041958041958
hi
0.11809815950920245
is
0.11538461538461538
vi
0.10852713178294573
reliable
true
passes
true
3
name
"Hunter"
dialogue
"Did you see anything… strange?"
detectedLang
"en"
heavyLang
"en"
scores
en
0.82174688057041
tl
0.6039603960396039
et
0.5337995337995338
da
0.494949494949495
no
0.48717948717948717
de
0.4609164420485175
nl
0.453551912568306
cs
0.44598337950138506
sq
0.4428969359331476
fr
0.42857142857142855
ms
0.41348973607038125
hr
0.3079584775086505
sk
0.3055555555555556
sl
0.3055555555555556
it
0.29328621908127206
pl
0.2805755395683453
ro
0.2727272727272727
sv
0.2647058823529412
lv
0.25925925925925924
is
0.250936329588015
lt
0.250936329588015
az
0.2277992277992278
tr
0.2277992277992278
reliable
true
passes
true
4
name
"Marie"
dialogue
"Oui, peut-être… J'ai quelque chose. Mais je ne sais pas… C’était bizarre."
detectedLang
"fr"
heavyLang
"fr"
scores
fr
0.8275862068965517
sq
0.4825708061002179
it
0.41574415744157445
sl
0.38788659793814434
hr
0.3741765480895916
en
0.33473389355742295
cs
0.3252840909090909
nl
0.31358381502890176
lt
0.2983751846381093
pt
0.2647058823529412
ca
0.2554858934169279
yo
0.2554858934169279
de
0.24960505529225907
sk
0.23387096774193547
lv
0.1782006920415225
ro
0.17677642980935876
tr
0.16520210896309317
pl
0.1607773851590106
fi
0.15630550621669628
ms
0.15026833631484796
hu
0.14721723518850988
reliable
true
passes
true
5
name
"Hunter"
dialogue
"Can I... see?"
detectedLang
"en"
heavyLang
"en"
scores
en
0.8287671232876712
et
0.7564935064935066
ca
0.7413793103448276
tr
0.6621621621621622
no
0.6590909090909091
az
0.6511627906976745
da
0.6495327102803737
hr
0.6495327102803737
sv
0.647887323943662
sq
0.6411483253588517
pl
0.6268656716417911
it
0.6052631578947368
cs
0.5833333333333334
ro
0.5454545454545454
yo
0.5370370370370371
tl
0.5
sk
0.4897959183673469
be
0.4827586206896552
hu
0.471830985915493
is
0.45652173913043476
lt
0.45652173913043476
lv
0.45652173913043476
uk
0.45652173913043476
vi
0.45652173913043476
reliable
true
passes
true
6
name
"Marie"
dialogue
"Là. Je ne sais pas ce que c’est..."
detectedLang
"fr"
heavyLang
"fr"
scores
fr
0.8339483394833949
ca
0.6024734982332155
sl
0.5722433460076045
it
0.563953488372093
hr
0.5579567779960707
yo
0.5273109243697479
ro
0.5192307692307693
sq
0.5140388768898488
cs
0.5044052863436124
nl
0.49095022624434387
lt
0.4730679156908665
es
0.4444444444444444
pt
0.43324937027707805
vi
0.4109947643979057
sk
0.39189189189189183
lv
0.31402439024390244
tr
0.29467084639498436
pl
0.2879746835443038
fi
0.281150159744409
ms
0.27184466019417475
hu
0.267100977198697
de
0.26229508196721313
reliable
true
passes
true
7
name
"Hunter"
dialogue
"Gotta go. Uh… stay safe."
detectedLang
"en"
heavyLang
"en"
scores
en
0.6724890829694322
it
0.5689655172413793
fi
0.42748091603053434
hu
0.4186046511627907
pl
0.4186046511627907
sv
0.33920704845814975
tl
0.33920704845814975
yo
0.33920704845814975
reliable
true
passes
true
8
name
"Marie"
dialogue
"Qu'est-ce qui se passe dans cette ville…?"
detectedLang
"fr"
heavyLang
"fr"
scores
fr
0.8397435897435898
da
0.6945010183299388
it
0.6753246753246753
no
0.6445497630331753
ca
0.5978552278820375
sv
0.5732574679943101
et
0.5184590690208668
de
0.5161290322580645
pt
0.5057660626029654
fi
0.494949494949495
ro
0.46524064171123
en
0.42857142857142855
nl
0.42748091603053434
es
0.39999999999999997
sq
0.37888198757763975
hu
0.3775933609958506
tl
0.33774834437086093
cs
0.3119266055045872
hr
0.3071593533487298
sl
0.3055555555555556
is
0.29906542056074764
yo
0.2628992628992629
tr
0.21465968586387435
ms
0.1780821917808219
reliable
true
passes
true