Run: 2026-02-25-01KJ94X8NY8BG1P84K69G5601T

Run Details

Model:
Gemini 3 Flash (Preview, Reasoning)
Test:
Language Writing
Scenario:
Character dialogue (Hindi) in a story
Duration:
12.6s
Cost:
$0.005509
Provider:
Google AI Studio

Evaluation summary

Score Evaluator
Details Link
58.33% 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: "Morning. Detective Hunter, Alborn PD. Mind if I have a moment of your time? I saw you out by the north orchards yesterday with that camera of yours." Selvi: "Oh! Shukr hai aap mil gaye. Mujhe kisi ko batana tha... wahan kuch bahut bura hua hai. Maine kuch dekha jo mujhe nahi dekhna chahiye tha." Hunter: "Whoa, slow down. I don’t speak... whatever that is. I’m looking for information on the disappearance of the girl from the apple festival. Did you see anything unusual near the treeline?" Selvi: "Maine us ladki ko dekha! Ek aadmi use ghasit raha tha pedon ke peeche. Maine photo lene ki koshish ki, par main bahut darr gayi thi. Mera haath abhi bhi kaanp raha hai, dekhiye!" Hunter: "Look, I’m not following a word of this. Is that Hindi? Listen, I just need a 'yes' or 'no.' Did you see a man in a red flannel jacket? Blue jeans?" Selvi: "Nahin, aap samajh nahi rahe hain! Voh aadmi... usne use pakad rakha tha. Maine bhagne ki koshish ki aur mera camera gir gaya. Photo nahi aa payi, par maine uski shakal dekhi!" Hunter: "Dammit. This is going nowhere fast. Can you show me the pictures on that thing? Just scroll back to yesterday afternoon. Let me see the screen." Selvi: "Camera kharab ho gaya hai! Jab main giri, toh vo pathar par laga. Main bas aapko batana chahti hoon ki vo ladki khatre mein hai. Hamein wahan wapas jaana chahiye!" Hunter: "I’ve got no idea what you’re stressed about, lady, but you look like you’ve seen a ghost. Just stay right here. I’m going to call the station and see if anyone on the force can translate." [Hunter’s phone rings loudly. He snaps it open immediately.] Hunter: "Hunter. What? Where? Near the old mill? Don't touch anything, I’m five minutes out." Hunter: "Look, I have to go. Duty calls. Just... stay in town, okay? I'll find you later. Don't go wandering off into the woods again." Selvi: "Rukiye! Aap jaa rahe hain? Par vo aadmi abhi bhi wahan ho sakta hai! Meri baat suniye!"
created_at"2026-02-25T00:59:12.969Z" 2/25/2026, 12:59:12 AM
model"google/gemini-3-flash-preview-20251217"
app_id182717
external_user(null)
streamedtrue
cancelledfalse
latency3654
moderation_latency(null)
generation_time12535
tokens_prompt315
tokens_completion962
native_tokens_prompt289
native_tokens_completion1788
native_tokens_completion_images0
native_tokens_reasoning1238
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"
usage0.0055085
router(null)
provider_responses
0
endpoint_id"6e82f652-2693-4b60-bbe0-7362d6e345f0"
is_byokfalse
latency3654
model_permaslug"google/gemini-3-flash-preview-20251217"
provider_name"Google AI Studio"
status200
api_type"completions"
id"gen-1771981152-pMv7s8IAcguvOHNx26Vr"
upstream_id(null)
total_cost0.0055085
cache_discount(null)
upstream_inference_cost0
provider_name"Google AI Studio"

Evaluation details

Result Evaluator Details Meta Data
58.33% Parse dialogue n/a
dialogue
0
name"Hunter"
dialogue"Morning. Detective Hunter, Alborn PD. Mind if I have a moment of your time? I saw you out by the north orchards yesterday with that camera of yours."
detectedLang"en"
heavyLang"en"
scores
en0.7959397237337799
tl0.6376811594202898
da0.5154075546719682
fr0.5020429009193054
no0.5005122950819673
ro0.49819866186309836
nl0.48358050847457623
it0.46013289036544847
ca0.45773081201334814
sv0.4497742663656885
cs0.3985194324491055
pt0.3917654398003743
de0.38446969696969696
es0.3741976893453145
sq0.3709677419354838
ms0.3512974051896207
yo0.3340163934426229
pl0.33356117566643884
sk0.32665745856353595
hr0.3229166666666667
hu0.2772424017790956
is0.25685975609756095
sl0.181360201511335
et0.17998317914213624
tr0.17442845046570704
az0.16737830913748933
lt0.12946428571428573
lv0.12003610108303249
vi0.12003610108303249
be0.06698564593301436
eu0.06698564593301436
fi0.06069364161849711
uk0.06069364161849711
reliabletrue
passestrue
1
name"Selvi"
dialogue"Oh! Shukr hai aap mil gaye. Mujhe kisi ko batana tha... wahan kuch bahut bura hua hai. Maine kuch dekha jo mujhe nahi dekhna chahiye tha."
detectedLang"tl"
heavyLang"tl"
scores
tl0.3755203996669442
eu0.37290969899665555
fi0.37027707808564225
ms0.3397887323943662
lv0.3042671614100185
sq0.2970946579194002
ca0.28977272727272724
sl0.2877492877492877
ro0.2809204218600192
fr0.27184466019417475
pt0.24698795180722888
es0.2401215805471125
et0.2401215805471125
yo0.2291880781089414
lt0.21465968586387435
az0.19786096256684493
it0.18566775244299674
en0.17853231106243156
vi0.1730981256890849
de0.1638795986622074
nl0.14285714285714285
da0.11971830985915494
sk0.10607866507747318
no0.09420289855072464
sv0.08980582524271845
hr0.08647990255785627
reliablefalse
passesfalse
2
name"Hunter"
dialogue"Whoa, slow down. I don’t speak... whatever that is. I’m looking for information on the disappearance of the girl from the apple festival. Did you see anything unusual near the treeline?"
detectedLang"en"
heavyLang"en"
scores
en0.8059822150363782
tl0.6575342465753424
nl0.5442461071021648
fr0.5249406175771971
no0.511002444987775
da0.50186799501868
et0.4562754870865428
sv0.4310099573257468
de0.42473633748801537
pt0.39607448414695523
it0.38303341902313626
yo0.37888198757763975
cs0.3736951983298539
ms0.3694167104571729
sq0.368753287743293
hr0.3637327677624602
es0.3442622950819672
ca0.329608938547486
sk0.3047508690614137
sl0.26784624771201954
pl0.26380368098159507
ro0.25880172946263125
fi0.23371647509578544
hu0.2263056092843327
lt0.21875
is0.2079207920792079
az0.20371599203715993
tr0.19463087248322147
lv0.18256130790190736
eu0.10044977511244377
hi0.06542056074766354
zh0.060297572435395456
be0.05511811023622047
uk0.0498812351543943
vi0.0498812351543943
ja0.04686258935663225
reliabletrue
passestrue
3
name"Selvi"
dialogue"Maine us ladki ko dekha! Ek aadmi use ghasit raha tha pedon ke peeche. Maine photo lene ki koshish ki, par main bahut darr gayi thi. Mera haath abhi bhi kaanp raha hai, dekhiye!"
detectedLang"en"
heavyLang"vi"
scores
en0.38816163915765506
fr0.3695014662756598
tl0.3095696852922286
lv0.2539902845246357
yo0.2359630419331912
ro0.23487544483985764
de0.2271746944644141
vi0.22494592645998557
sq0.22382671480144403
et0.22044960116026108
tr0.2147552958363769
sl0.20605612998522896
fi0.19955323901712585
da0.15553809897879026
sv0.15553809897879026
cs0.15086887835703003
az0.12672623883021933
ms0.11522633744855967
hu0.10117056856187291
eu0.08898305084745763
it0.0851063829787234
nl0.0851063829787234
lt0.0811965811965812
hi0.06684027777777778
ca0.06521739130434782
es0.06521739130434782
pt0.06521739130434782
reliablefalse
passesfalse
4
name"Hunter"
dialogue"Look, I’m not following a word of this. Is that Hindi? Listen, I just need a 'yes' or 'no.' Did you see a man in a red flannel jacket? Blue jeans?"
detectedLang"en"
heavyLang"en"
scores
en0.7967020555681048
tl0.6976822304333221
sv0.47856315179606024
nl0.4721407624633431
da0.43467336683417085
et0.4121489222730242
it0.4102228047182176
de0.399599733155437
no0.3923024983119514
yo0.36930623686054664
is0.3511175198269647
hu0.33677229182019164
ro0.3253373313343328
hr0.3082244427363567
pl0.29797191887675506
cs0.29577464788732394
sl0.29577464788732394
sq0.2834394904458599
fr0.281150159744409
ca0.2805755395683453
lv0.2665036674816626
tr0.24496644295302009
es0.2366412213740458
sk0.2248062015503876
fi0.21807124239791487
lt0.20983318700614575
az0.19427036705461057
pt0.18404351767905713
eu0.172033118675253
ms0.13294797687861273
vi0.1287512100677638
hi0.07975460122699386
be0.07216494845360825
uk0.06542056074766354
reliabletrue
passestrue
5
name"Selvi"
dialogue"Nahin, aap samajh nahi rahe hain! Voh aadmi... usne use pakad rakha tha. Maine bhagne ki koshish ki aur mera camera gir gaya. Photo nahi aa payi, par maine uski shakal dekhi!"
detectedLang"tl"
heavyLang"vi"
scores
tl0.5067372473532242
fr0.41361556064073224
en0.3895175699821322
ms0.31208053691275167
eu0.3050847457627119
lv0.29938482570061514
ro0.2857142857142857
sv0.25992779783393505
no0.2457689477557027
et0.206656346749226
fi0.206656346749226
da0.19984387197501954
it0.1865079365079365
de0.1760450160771704
az0.13209144792548688
nl0.1298811544991511
sl0.1298811544991511
hu0.10558464223385688
sq0.10245183887915937
yo0.0992970123022847
tr0.0921169176262179
ca0.09050576752440107
is0.09050576752440107
es0.08807829181494661
lt0.08318425760286224
cs0.07906558849955077
hi0.06987295825771324
reliablefalse
passesfalse
6
name"Hunter"
dialogue"Dammit. This is going nowhere fast. Can you show me the pictures on that thing? Just scroll back to yesterday afternoon. Let me see the screen."
detectedLang"en"
heavyLang"en"
scores
en0.8004434589800443
tl0.6165317426501917
nl0.47368421052631576
no0.438552713661884
et0.40750493745885447
sv0.37586685159500693
da0.36395759717314485
ca0.3576017130620985
fr0.3430656934306569
yo0.3278566094100075
cs0.3055555555555556
sl0.25864909390444807
fi0.2555831265508685
sq0.2524916943521595
sk0.23922231614539302
es0.2194275802254987
de0.2119089316987741
pt0.2112182296231376
ms0.21052631578947367
pl0.19571045576407506
lv0.18552036199095023
hr0.1847826086956522
hu0.14611005692599618
ro0.14448669201520914
lt0.10358565737051792
is0.09274193548387095
hi0.08536585365853659
it0.0816326530612245
az0.07881269191402251
tr0.07881269191402251
reliabletrue
passestrue
7
name"Selvi"
dialogue"Camera kharab ho gaya hai! Jab main giri, toh vo pathar par laga. Main bas aapko batana chahti hoon ki vo ladki khatre mein hai. Hamein wahan wapas jaana chahiye!"
detectedLang"ms"
heavyLang"tlh"
scores
ms0.40293937538273117
fr0.37540038436899426
ca0.3257261410788381
it0.30851063829787234
tl0.30057388809182206
fi0.2893586005830903
es0.25968109339407747
ro0.2494226327944573
en0.22310756972111553
de0.22
sk0.214975845410628
et0.19950738916256158
pt0.19950738916256158
az0.1948802642444261
sl0.18682235195996663
no0.18614357262103506
sv0.15217391304347824
vi0.13869257950530034
lv0.13716814159292035
hu0.1104014598540146
yo0.10386029411764705
is0.10303587856485741
cs0.09805735430157261
tr0.09638554216867469
lt0.09302325581395349
da0.08707865168539326
eu0.08622305529522024
nl0.07231208372978117
hr0.06787762906309751
reliablefalse
passesfalse
8
name"Hunter"
dialogue"I’ve got no idea what you’re stressed about, lady, but you look like you’ve seen a ghost. Just stay right here. I’m going to call the station and see if anyone on the force can translate."
detectedLang"en"
heavyLang"en"
scores
en0.8187601716230211
tl0.6056020605280102
fr0.4673913043478261
no0.4509188704616764
yo0.3850401606425703
ca0.38037430450177034
sv0.37017994858611825
da0.36823104693140796
hr0.3559411146161935
et0.35219460602855635
sk0.3417517463729178
cs0.32766190998902306
pl0.3269230769230769
it0.32245575221238937
de0.3106359032076534
sq0.3106359032076534
lv0.2947610823258492
ro0.26514697060587883
es0.2611580217129071
nl0.2602657004830918
sl0.2484662576687117
lt0.2400744416873449
tr0.2290748898678414
pt0.22615287428932404
ms0.21524663677130043
fi0.19407894736842105
az0.18005354752342703
vi0.11424439624005785
hu0.11296162201303403
eu0.10058737151248165
is0.09860191317144959
hi0.0641711229946524
zh0.059139784946236555
reliabletrue
passestrue
9
name"Hunter"
dialogue"Hunter. What? Where? Near the old mill? Don't touch anything, I’m five minutes out."
detectedLang"en"
heavyLang"en"
scores
en0.8175182481751825
tl0.5696400625978091
fr0.4393476044852191
sq0.4073275862068965
de0.359720605355064
ms0.3325242718446602
no0.3037974683544304
da0.29214929214929214
sv0.2838541666666667
nl0.2763157894736842
ca0.27344782034346105
et0.20634920634920634
sl0.19236417033773862
it0.19117647058823528
ro0.18998527245949925
pt0.1863905325443787
yo0.1628614916286149
cs0.12
sk0.12
fi0.11575562700964631
es0.1085899513776337
hr0.1085899513776337
lt0.10277324632952692
pl0.10277324632952692
tr0.09688013136288999
reliabletrue
passestrue
10
name"Hunter"
dialogue"Look, I have to go. Duty calls. Just... stay in town, okay? I'll find you later. Don't go wandering off into the woods again."
detectedLang"en"
heavyLang"en"
scores
en0.8088197807800154
nl0.5598591549295775
tl0.5446265938069217
da0.5401594114040466
de0.4688385269121813
no0.4592645998558039
sv0.4545454545454546
pl0.41360437842064113
yo0.3980738362760835
cs0.2951127819548872
ca0.2877492877492877
sl0.2682926829268293
it0.258893280632411
hr0.2537313432835821
ms0.24623115577889448
ro0.23625254582484725
sq0.2203742203742204
sk0.21383647798742136
lv0.19786096256684493
lt0.18211559432933477
et0.18032786885245902
fr0.1657397107897664
az0.1638795986622074
es0.11137440758293839
hi0.09200968523002422
pt0.08647990255785627
be0.08536585365853659
tr0.08536585365853659
hu0.08200734394124846
is0.07749077490774908
uk0.07749077490774908
vi0.07749077490774908
reliabletrue
passestrue
11
name"Selvi"
dialogue"Rukiye! Aap jaa rahe hain? Par vo aadmi abhi bhi wahan ho sakta hai! Meri baat suniye!"
detectedLang"tr"
heavyLang"la"
scores
tr0.3840985442329227
sk0.3268053855569155
ms0.3064312736443884
it0.2724867724867725
no0.2537313432835821
sq0.24760601915184677
sv0.23504867872044508
vi0.22206506364922204
lv0.2198581560283688
fr0.207492795389049
eu0.2005813953488372
tl0.19236417033773862
ca0.1791044776119403
cs0.16158536585365854
et0.16030534351145037
lt0.15384615384615385
da0.14463452566096424
fi0.13793103448275862
is0.12140575079872205
reliablefalse
passesfalse