NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2024-08-09-01J4W2CF5A0PWCA1GH2F0QAGR0
Run Details
Model:
Llama 3.1 70B
Test:
Language Writing
Scenario:
Character dialogue (Hindi) in a story
Duration:
11.7s
Cost:
$0.000500
Provider:
Novita
Evaluation summary
Score
Evaluator
Details Link
57.14%
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="Selvi"> Selvi is a 30-year old college student on a study abroad program. Born in India, she is fluent in Hindi 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 Hindi, 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: Selvi 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 Hindi, and Hunter looks confused. They talk for a while, until Hunter gets a call and has to leave. Return the dialogue between Selvi and Hunter as a voice sheet: Selvi: "..." Hunter: "..."
Model response
Messages
Meta
AI
Here is the scene: Selvi sat at a small café in the heart of Alborn, sipping on a cup of coffee and examining her camera lens. She looked up to see a man with a rugged jawline and a stern expression approaching her. Hunter flashed his detective badge. "Excuse me, miss. I'm Detective Hunter. I'm investigating some strange occurrences in town and I understand you may have some information." Selvi's eyes widened, and she responded in a worried tone, "Pichhle din ek ajeeb si baat hui thi. Maine kuch dekha jo mujhe bahut pareshan kar raha hai." (Yesterday, something strange happened. I saw something that's been troubling me a lot.) Hunter's expression changed from curious to confused. "I'm sorry, miss...?" Selvi tried again, this time slower, hoping he would understand, "Maine ek ladka dekha, jo ekdum gaayab ho gaya. Usne kabhi bhi kuch nahi kaha." (I saw a boy, who suddenly disappeared. He didn't say a word.) Hunter's confusion deepened. "I'm afraid I don't understand what you're saying, miss. Do you speak English?" Selvi shook her head, her eyes welling up with tears. "Nahin... mujhe angrezi nahi aati." (No... I don't know English.) Hunter's face softened slightly, and he tried to reassure her, "It's okay, I'll figure out a way to communicate with you. Can you tell me what's wrong?" But his face showed his struggle to understand what she was saying. Just then, his phone rang. He answered, listening for a moment before apologizing, "I have to go, miss. If you don't mind, can I come back and talk to you later with a translator?" He quickly took down her name and number and rushed out of the café. As he left, Selvi let out a sigh, her eyes returning to her camera. She flipped through the pictures, stopping on one of the town's apple festival. But her mind was elsewhere, on the strange occurrence that had left her shaken.
id
"gen-USDUOJAmf8xMjHGFqQ2kuEvQEdPg"
total_cost
0.00050022
upstream_id
"cmpl-604915dd6f054e41b796a0927731a80f"
created_at
"2024-08-09T17:18:47.718727+00:00"
8/9/2024, 5:18:47 PM
model
"meta-llama/llama-3.1-70b-instruct"
app_id
182717
streamed
false
cancelled
false
provider_name
"Novita"
latency
11524
moderation_latency
(null)
generation_time
0
finish_reason
"stop"
tokens_prompt
324
tokens_completion
460
native_tokens_prompt
278
native_tokens_completion
457
num_media_prompt
(null)
num_media_completion
(null)
origin
"https://ncbench.com/"
usage
0.00050022
Evaluation details
Result
Evaluator
Details
Meta Data
57.14%
Parse dialogue
n/a
dialogue
0
name
"Hunter"
dialogue
"Excuse me, miss. I'm Detective Hunter. I'm investigating some strange occurrences in town and I understand you may have some information."
detectedLang
"en"
heavyLang
"en"
scores
en
0.802349220691213
tl
0.6787812041116007
da
0.631888935633151
fr
0.6167323696890057
no
0.5950948634891254
de
0.5892018779342723
nl
0.5886224729666196
sv
0.5801343570057582
ro
0.5668316831683169
it
0.558303886925795
pt
0.49451184286539573
ca
0.48377581120943947
sq
0.48102016607354686
sl
0.47352587244283995
cs
0.4433842239185751
es
0.4347545219638243
hr
0.4220607661822986
yo
0.3829337094499295
sk
0.3754461099214847
pl
0.37410586552217456
ms
0.36270939548434084
lv
0.36084733382030676
et
0.3484735666418466
az
0.33962264150943394
lt
0.31533646322378717
is
0.28101889893179954
tr
0.22771403353927622
fi
0.19798350137488543
vi
0.15130940834141612
eu
0.13280475718533202
hu
0.12935323383084577
zh
0.08088235294117647
be
0.07407407407407407
uk
0.06716417910447761
reliable
true
passes
true
1
name
"Selvi"
dialogue
"Pichhle din ek ajeeb si baat hui thi. Maine kuch dekha jo mujhe bahut pareshan kar raha hai."
detectedLang
"ro"
heavyLang
"de"
scores
ro
0.43868739205526774
fr
0.4268077601410935
tl
0.39422180801491147
et
0.37739463601532564
sl
0.364613880742913
fi
0.35323383084577115
sq
0.3340163934426229
vi
0.32432432432432434
it
0.31650893796004204
lv
0.31362196409714893
lt
0.3010752688172043
ca
0.2988133764832794
en
0.2655367231638418
yo
0.24855491329479767
tr
0.2459396751740139
da
0.2388758782201405
es
0.23076923076923078
no
0.19753086419753088
sv
0.18952618453865336
de
0.18444165621079048
sk
0.1591203104786546
cs
0.15474642392717813
nl
0.13333333333333333
eu
0.11444141689373298
pt
0.11202185792349727
reliable
false
passes
false
2
name
"Hunter"
dialogue
"I'm sorry, miss...?"
detectedLang
"en"
heavyLang
"en"
scores
en
0.7530864197530864
tl
0.6666666666666666
it
0.5
pt
0.5
reliable
true
passes
true
3
name
"Selvi"
dialogue
"Maine ek ladka dekha, jo ekdum gaayab ho gaya. Usne kabhi bhi kuch nahi kaha."
detectedLang
"fi"
heavyLang
"la"
scores
fi
0.34210526315789475
eu
0.3217054263565891
ca
0.284741144414169
fr
0.2837653478854025
lv
0.28082191780821925
tl
0.2698191933240611
no
0.2647058823529412
en
0.2369186046511628
et
0.2369186046511628
de
0.21875
ro
0.21875
tr
0.1758241758241758
sk
0.17452830188679244
lt
0.1732283464566929
cs
0.16798732171156894
it
0.16798732171156894
sl
0.16798732171156894
da
0.16267942583732056
ms
0.16
nl
0.16
sq
0.15322580645161288
yo
0.14910858995137766
reliable
false
passes
false
4
name
"Hunter"
dialogue
"I'm afraid I don't understand what you're saying, miss. Do you speak English?"
detectedLang
"en"
heavyLang
"en"
scores
en
0.7998475028593214
no
0.49952335557673977
tl
0.49904580152671757
da
0.493731918997107
sv
0.4450317124735729
cs
0.4324324324324324
pl
0.3979357798165138
hr
0.38668224299065423
sk
0.3581907090464548
de
0.3354430379746835
sq
0.3234536082474227
ro
0.3101182654402102
nl
0.2532005689900427
vi
0.24892703862660948
tr
0.2446043165467626
yo
0.2391304347826087
pt
0.2290748898678414
ca
0.2081447963800905
is
0.20694864048338368
sl
0.18604651162790697
it
0.1796875
ms
0.12060301507537688
be
0.11764705882352942
hu
0.11317567567567569
az
0.10714285714285714
et
0.10714285714285714
lt
0.10714285714285714
lv
0.10714285714285714
uk
0.10714285714285714
reliable
true
passes
true
5
name
"Selvi"
dialogue
"Nahin... mujhe angrezi nahi aati."
detectedLang
"eu"
heavyLang
"tl"
scores
eu
0.609375
tl
0.5359628770301624
ro
0.3442622950819672
no
0.3197278911564626
da
0.3055555555555556
de
0.2805755395683453
reliable
true
passes
false
6
name
"Hunter"
dialogue
"It's okay, I'll figure out a way to communicate with you. Can you tell me what's wrong?"
detectedLang
"en"
heavyLang
"en"
scores
en
0.8091603053435115
tl
0.6301277740416947
fr
0.5480690221857025
it
0.5084897229669348
ro
0.4670542635658915
no
0.43064182194616984
nl
0.4155154091392136
es
0.40476190476190477
pt
0.4008714596949891
ca
0.3962678375411636
cs
0.38271604938271603
sk
0.37570942111237227
da
0.36342592592592593
ms
0.36194895591647336
sl
0.35217903415783275
sq
0.35064935064935066
yo
0.3176178660049628
pl
0.3037974683544304
hr
0.2930591259640103
et
0.29214929214929214
vi
0.289405684754522
az
0.2819843342036553
de
0.2791612057667104
lv
0.25876010781671155
lt
0.25271739130434784
hu
0.22096317280453256
tr
0.21875
fi
0.2005813953488372
sv
0.1935483870967742
is
0.12
eu
0.11290322580645161
reliable
true
passes
true