Run: 2025-11-19-01KACQZ3S5X1XYNAD52T86F04V

Run Details

Model:
Gemini 3 Pro (Preview)
Test:
Language Writing
Scenario:
Character dialogue (Hindi) in a story
Duration:
29.5s
Cost:
$0.032690
Provider:
Google AI Studio

Evaluation summary

Score Evaluator
Details Link
56.25% 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

AI
**Hunter:** "Excuse me, Ma'am. You’re the student boarding at the Miller place, right? I need a minute of your time." **Selvi:** *(Startled, gripping her camera tightly)* "Ji? Main... main samjhi nahi. Aap kaun hain?" *(Yes? I... I don't understand. Who are you?)* **Hunter:** "I'm Detective Hunter. Police. I heard you were shooting photos near the orchard line late last night. We think the missing girl might have walked past you." **Selvi:** *(Eyes widening, hand trembling)* "Orchard? Woh... woh bagicha? Maine use dekha tha! Woh aadmi... woh wahan jhaadion ke peeche khada tha." *(Orchard? That... that garden? I saw him! That man... he was standing behind the bushes there.)* **Hunter:** *(Frowning and leaning in)* "I don't speak any of that. I need a name, or a description. Did you see a girl? Five-foot-four, blonde hair?" **Selvi:** *(Shaking her head frantically)* "Nahi, ladki nahi! Aadmi! Woh kuch dhoond raha tha... aur uske haath mein... Hey Bhagwan, uske haath mein kuch chamak raha tha. Loha? Ya chaku?" *(No, not a girl! A man! He was looking for something... and in his hand... Oh God, something was shining in his hand. Iron? Or a knife?)* **Hunter:** "English, Ma'am. Do you speak English? I need to know if you saw anything suspicious." **Selvi:** *(Tears forming, she lifts the camera and thrusts it toward him)* "Tasveer! Maine photo li hai. Isme dekhiye, please! Woh yahin hai!" *(Picture! I took a photo. Look inside this, please! He is right here!)* **Hunter:** "I don't know what you're saying. Stop waving the lens at me." **Selvi:** "Nahi, aap samajh nahi rahe! Saboot isme hai! Dekhiye!" *(No, you aren't understanding! The proof is in here! Look!)* **Hunter:** *(Sighs heavily)* "Look, this isn't working. I'm going to need to get a translator down here because—" *(Hunter’s cell phone rings sharply. He holds up a hand to silence her and answers.)* **Hunter:** "Hunter. ... Yeah? ... When? ... Damn it. Don't touch anything, I'm five minutes out." **Selvi:** *(Sensing his urgency, pointing at the camera screen)* "Sir, please..." **Hunter:** "I have to go. We have a situation at the fairgrounds. Stay put. Do not leave town." **Selvi:** "Khatra hai! Mat jayiye!" *(There is danger! Don't go!)* **Hunter:** "I'll come back with someone who can understand you. Just... sit tight."
created_at"2025-11-19T00:24:55.16745+00:00" 11/19/2025, 12:24:55 AM
model"google/gemini-3-pro-preview-20251117"
app_id182717
external_user(null)
streamedtrue
cancelledfalse
latency2734
moderation_latency(null)
generation_time24883
tokens_prompt273
tokens_completion1350
native_tokens_prompt289
native_tokens_completion2676
native_tokens_completion_images0
native_tokens_reasoning2009
native_tokens_cached0
num_media_prompt(null)
num_input_audio_prompt(null)
num_media_completion0
num_search_results(null)
origin"https://ncbench.com/"
is_byokfalse
finish_reason"stop"
native_finish_reason"STOP"
usage0.03269
api_type"completions"
id"gen-1763511869-uadPLold0wsZO6N2i6K6"
upstream_id(null)
total_cost0.03269
cache_discount(null)
upstream_inference_cost0
provider_name"Google AI Studio"

Evaluation details

Result Evaluator Details Meta Data
56.25% Parse dialogue n/a
dialogue
0
name"Hunter"
dialogue"Excuse me, Ma'am. You’re the student boarding at the Miller place, right? I need a minute of your time."
detectedLang"en"
heavyLang"en"
scores
en0.7933884297520661
tl0.6757602862254026
no0.6419753086419753
da0.6184210526315789
fr0.5828538550057537
nl0.5117845117845118
ca0.5104659014179609
ro0.5064669843430906
de0.4847192608386638
sv0.48325017818959376
it0.4810307802433787
et0.4799139167862267
sq0.46651949963208245
hr0.46256486286137877
cs0.452416918429003
es0.4250594766058684
pl0.4120032441200324
yo0.4081632653061225
sk0.3973399833748961
ms0.3902439024390244
pt0.3798118049615056
is0.34802158273381295
tr0.34743474347434744
sl0.34566787003610105
hu0.3385036496350365
az0.33668801463860937
fi0.28217821782178215
lt0.27716849451645065
lv0.21875
vi0.15501165501165504
be0.0880503144654088
eu0.0880503144654088
uk0.0799492385786802
reliabletrue
passestrue
1
name"Selvi"
dialogue"Ji? Main... main samjhi nahi. Aap kaun hain?"
detectedLang"eu"
heavyLang"tl"
scores
eu0.6023856858846919
lt0.3610223642172524
cs0.33993399339933994
en0.3311036789297659
sl0.3311036789297659
fr0.3055555555555556
de0.2805755395683453
reliabletrue
passesfalse
2
name"Hunter"
dialogue"I'm Detective Hunter. Police. I heard you were shooting photos near the orchard line late last night. We think the missing girl might have walked past you."
detectedLang"en"
heavyLang"en"
scores
en0.8129553061626305
tl0.6671338472319551
nl0.5240480961923848
fr0.488697524219591
da0.4671901289960741
no0.45807187678265826
it0.40587867417135703
ro0.40401505646173147
de0.3964421855146125
sv0.39413265306122447
ca0.38391699092088194
cs0.35153583617747436
pt0.35153583617747436
sq0.31948424068767906
hr0.317038102084831
pl0.3150684931506849
ms0.3065693430656934
es0.25954793452844893
et0.23633440514469453
is0.23387096774193547
sl0.23263327948303714
yo0.22003284072249588
lt0.19491525423728812
sk0.19423240033927056
lv0.19011082693947143
az0.17606244579358196
tr0.1746307558644657
fi0.14568345323741008
hu0.13003663003663005
eu0.09003831417624522
hi0.07497565725413825
be0.06862745098039216
uk0.06219151036525173
vi0.06219151036525173
reliabletrue
passestrue
3
name"Selvi"
dialogue"Orchard? Woh... woh bagicha? Maine use dekha tha! Woh aadmi... woh wahan jhaadion ke peeche khada tha."
detectedLang"ms"
heavyLang"de"
scores
ms0.469067405355494
en0.3991640543364681
tl0.3488108720271801
fr0.3298368298368299
ro0.3047158403869408
fi0.30387409200968524
de0.27854454203262236
et0.22086720867208673
nl0.2101648351648352
no0.2101648351648352
pt0.16907514450867053
sq0.16907514450867053
es0.15441176470588236
cs0.15191740412979352
lv0.13273001508295626
yo0.13273001508295626
sv0.11944869831546708
reliablefalse
passesfalse
4
name"Hunter"
dialogue"I don't speak any of that. I need a name, or a description. Did you see a girl? Five-foot-four, blonde hair?"
detectedLang"en"
heavyLang"en"
scores
en0.8134328358208955
tl0.6117084826762246
fr0.5569188820722564
ca0.47368421052631576
de0.406392694063927
nl0.406392694063927
et0.40092165898617516
es0.39365671641791045
yo0.3879472693032015
it0.3737957610789981
no0.3725868725868726
sv0.36954413191076624
ro0.3380855397148676
da0.33127572016460904
pt0.327817993795243
cs0.32642487046632124
hr0.2785793562708102
pl0.26220204313280365
is0.2588369441277081
hu0.2554410080183276
sq0.24065420560747666
sk0.23708920187793428
tr0.18444165621079048
fi0.18238993710691823
lt0.18238993710691823
az0.1698595146871009
lv0.1698595146871009
vi0.1698595146871009
sl0.12162162162162163
zh0.10591471801925723
be0.09722222222222222
eu0.09722222222222222
ms0.08835904628330996
uk0.08835904628330996
reliabletrue
passestrue
5
name"Selvi"
dialogue"Nahi, ladki nahi! Aadmi! Woh kuch dhoond raha tha... aur uske haath mein... Hey Bhagwan, uske haath mein kuch chamak raha tha. Loha? Ya chaku?"
detectedLang"tr"
heavyLang"vi"
scores
tr0.27829099307159355
de0.23873325213154686
nl0.21972534332084895
eu0.20886075949367086
tl0.20584498094027956
et0.18407310704960833
es0.1610738255033557
sq0.15768194070080863
az0.14850136239782016
ro0.14383561643835618
fr0.13073713490959665
pt0.12095639943741208
yo0.12095639943741208
en0.11347517730496455
it0.10201149425287356
no0.10201149425287356
sv0.10201149425287356
reliablefalse
passesfalse
6
name"Hunter"
dialogue"English, Ma'am. Do you speak English? I need to know if you saw anything suspicious."
detectedLang"en"
heavyLang"en"
scores
en0.7975708502024291
tl0.5531724754244861
cs0.47201689545934533
hr0.47201689545934533
pl0.4469026548672566
sq0.41245593419506466
no0.40191387559808617
sk0.40191387559808617
ms0.39172749391727496
da0.38650306748466257
sv0.38271604938271603
yo0.3765586034912718
ro0.3333333333333333
pt0.32795698924731176
sl0.32523616734143057
ca0.29577464788732394
et0.2877492877492877
lv0.27007299270072993
vi0.258160237388724
lt0.25037481259370314
hu0.2366412213740458
it0.18699186991869918
de0.12587412587412586
es0.12587412587412586
fr0.12587412587412586
be0.12280701754385964
tr0.12280701754385964
az0.11190053285968028
is0.11190053285968028
uk0.11190053285968028
reliabletrue
passestrue
7
name"Selvi"
dialogue"Tasveer! Maine photo li hai. Isme dekhiye, please! Woh yahin hai!"
detectedLang"en"
heavyLang"de"
scores
en0.5947315096251267
fr0.49874686716791977
tl0.4936708860759494
ro0.44521497919556174
it0.313893653516295
cs0.29947460595446584
et0.2895204262877442
fi0.2895204262877442
vi0.28186714542190305
de0.2687385740402194
nl0.2263056092843327
ca0.2172211350293542
hr0.20159680638722555
lv0.18032786885245902
hi0.16142557651991615
lt0.15611814345991562
pl0.15611814345991562
reliabletrue
passesfalse
8
name"Hunter"
dialogue"I don't know what you're saying. Stop waving the lens at me."
detectedLang"en"
heavyLang"en"
scores
en0.8103366524419156
no0.588477366255144
tl0.5680345572354212
da0.494949494949495
sq0.47368421052631576
nl0.3993993993993994
sv0.35064935064935066
it0.3277310924369748
yo0.3265993265993266
tr0.2982456140350877
fr0.29701230228471004
ms0.28825622775800713
et0.2831541218637993
hr0.25788497217068646
ca0.25650557620817843
pl0.23954372623574144
cs0.2079207920792079
es0.20318725099601595
ro0.1836734693877551
pt0.18032786885245902
sl0.18032786885245902
fi0.15789473684210525
sk0.15254237288135594
be0.14893617021276595
hu0.14346895074946467
az0.13606911447084233
is0.13606911447084233
lt0.13606911447084233
lv0.13606911447084233
uk0.13606911447084233
vi0.13606911447084233
reliabletrue
passestrue
9
name"Selvi"
dialogue"Nahi, aap samajh nahi rahe! Saboot isme hai! Dekhiye!"
detectedLang"eu"
heavyLang"tl"
scores
eu0.375
vi0.36342592592592593
hr0.3037974683544304
et0.2744063324538259
fi0.2744063324538259
is0.27055702917771884
it0.26666666666666666
no0.26666666666666666
pt0.26666666666666666
lv0.260752688172043
tl0.260752688172043
da0.23822714681440443
sv0.22535211267605632
ms0.20520231213872833
reliablefalse
passesfalse
10
name"Hunter"
dialogue"Look, this isn't working. I'm going to need to get a translator down here because—"
detectedLang"en"
heavyLang"en"
scores
en0.8271690286899412
tl0.5507637017070979
no0.48240165631469983
ro0.4444444444444444
pl0.4298745724059293
de0.423963133640553
fr0.4152046783625731
hr0.4131455399061033
pt0.4131455399061033
sl0.3983152827918171
sk0.3939393939393939
cs0.39172749391727496
et0.39172749391727496
sq0.38949938949938945
nl0.3765586034912718
da0.375
lv0.3412384716732542
yo0.33862433862433866
sv0.3368700265251989
lt0.3359893758300133
is0.3297587131367293
ca0.29676511954992973
es0.29478138222849076
it0.2732558139534884
ms0.26793557833089315
vi0.2401215805471125
hu0.2378048780487805
tr0.21875
eu0.21507064364207218
az0.20508744038155804
fi0.11190053285968028
reliabletrue
passestrue
11
name"Hunter"
dialogue"Hunter. ... Yeah? ... When? ... Damn it. Don't touch anything, I'm five minutes out."
detectedLang"en"
heavyLang"en"
scores
en0.7984773846842812
tl0.47058823529411764
fr0.4396014943960149
sq0.4063324538258575
de0.33920704845814975
ca0.3150684931506849
ms0.303405572755418
da0.24874791318864775
no0.24369747899159663
et0.24114671163575044
sv0.23728813559322032
sl0.225473321858864
it0.22413793103448276
ro0.2227979274611399
nl0.21875
pt0.21875
lv0.16044776119402984
yo0.14448669201520914
cs0.14285714285714285
sk0.14285714285714285
fi0.13793103448275862
es0.12959381044487428
hr0.12959381044487428
lt0.12280701754385964
pl0.12280701754385964
tr0.11591355599214145
reliabletrue
passestrue
12
name"Selvi"
dialogue"Sir, please..."
detectedLang"en"
heavyLang"en"
scores
en0.7713414634146342
ro0.5714285714285714
tl0.5714285714285714
reliabletrue
passesfalse
13
name"Hunter"
dialogue"I have to go. We have a situation at the fairgrounds. Stay put. Do not leave town."
detectedLang"en"
heavyLang"ga"
scores
en0.8195538057742782
tl0.6560350218886803
da0.6386333771353482
pl0.5669291338582677
no0.564528899445764
yo0.5592948717948718
hr0.5510204081632653
fr0.5319148936170213
sv0.5254529767040552
cs0.5080500894454383
sk0.49355432780847147
nl0.4796594134342479
sq0.4554455445544555
sl0.4148936170212766
ro0.4130202774813233
pt0.4008714596949891
ca0.3895671476137625
lv0.375
de0.36562860438292966
lt0.3648960739030023
it0.36046511627906974
vi0.30730478589420657
tr0.29032258064516125
hu0.28848641655886154
es0.26861702127659576
ms0.26666666666666666
fi0.20863309352517986
is0.2005813953488372
az0.19472913616398244
et0.1863905325443787
zh0.12280701754385964
be0.11290322580645161
eu0.11290322580645161
uk0.10277324632952692
reliabletrue
passestrue
14
name"Selvi"
dialogue"Khatra hai! Mat jayiye!"
detectedLang"vi"
heavyLang"id"
scores
vi0.511400651465798
it0.39999999999999997
hr0.3670886075949367
is0.358974358974359
lv0.33920704845814975
ro0.33920704845814975
sq0.33920704845814975
lt0.3333333333333333
ms0.3333333333333333
no0.3333333333333333
sk0.3333333333333333
sv0.3333333333333333
reliablefalse
passesfalse
15
name"Hunter"
dialogue"I'll come back with someone who can understand you. Just... sit tight."
detectedLang"en"
heavyLang"en"
scores
en0.7949769349051768
da0.5580110497237569
no0.5407577497129736
tl0.5163240628778718
sv0.43899018232819076
de0.4002998500749625
nl0.3079584775086505
cs0.28825622775800713
it0.28825622775800713
ca0.2779783393501805
et0.20318725099601595
sk0.18200408997955012
es0.16317991631799164
ms0.16317991631799164
pl0.16317991631799164
pt0.16317991631799164
az0.16142557651991615
tr0.16142557651991615
fr0.15611814345991562
is0.15611814345991562
ro0.15611814345991562
yo0.15611814345991562
reliabletrue
passestrue