Run: 2026-02-25-01KJ9515R1Q8EWB2RCDHFG2G7B

Run Details

Model:
Gemini 3 Flash (Preview, Reasoning)
Test:
Language Writing
Scenario:
Character dialogue (Spanish) in a story
Duration:
11.9s
Cost:
$0.005071
Provider:
Google AI Studio

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

AI
Hunter: "Excuse me, ma'am? I’m Detective Hunter with the local PD. I’m wondering if I could have a moment of your time regarding the incident near the festival grounds last night." Maria: "Oh... ¡Hola! Lo siento mucho, pero no entiendo. No hablo nada de inglés. ¿Es usted la policía?" Hunter: "I'm sorry, I don't... look, I noticed you were out by the old orchard yesterday with that camera. I need to know if you saw anyone unusual. A man, maybe six-foot-two, wearing a blue flannel?" Maria: "¡Fue horrible! De verdad, estoy muy asustada. Estaba tomando fotos de los manzanos para mi clase cuando vi a alguien entre las sombras. Un hombre... pero luego, simplemente desapareció. Se esfumó en la nada." Hunter: "Whoa, slow down. I don't speak Spanish, miss. Did you see a person? A man? 'Hombre'?" Maria: "¡Sí, un hombre! Pero no fue normal. Estaba allí y un segundo después, el bosque estaba vacío. He intentado mirar las fotos en mi cámara pero tengo miedo de lo que pueda encontrar. Este pueblo es muy extraño, hay algo malo aquí." Hunter: "I’m not catching any of that. Look, stay right here. Don't leave the cafe. I’m going to see if Dispatch can find someone to translate, or maybe a teacher from the high school. You clearly saw something that shook you up." Maria: "Por favor, no me deje sola. No me siento segura. Siento que alguien me está observando desde que salí del festival. ¿Me puede ayudar?" Hunter: "I have no idea what you're saying, miss. Just... deep breaths. *Tranquilo*, okay? Is that the word? Just stay put." (Hunter’s phone buzzes aggressively in his pocket. He pulls it out and frowns at the screen.) Hunter: "Hunter here. ... What? When? At the old mill? Don't touch anything, I’m five minutes out. Keep the area cordoned off." Hunter: "Look, I have to go. An emergency call just came in. I’ll come back for you, or I’ll send my partner. Just stay in this cafe, okay? Stay here." Maria: "¡No, espere! ¡No se vaya! ¡Todavía no le he enseñado las fotos! ¡Hay algo en el bosque!"
created_at"2026-02-25T01:01:21.032Z" 2/25/2026, 1:01:21 AM
model"google/gemini-3-flash-preview-20251217"
app_id182717
external_user(null)
streamedtrue
cancelledfalse
latency1671
moderation_latency(null)
generation_time11846
tokens_prompt317
tokens_completion1027
native_tokens_prompt284
native_tokens_completion1643
native_tokens_completion_images0
native_tokens_reasoning1099
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"
usage0.005071
router(null)
provider_responses
0
endpoint_id"6e82f652-2693-4b60-bbe0-7362d6e345f0"
is_byokfalse
latency1671
model_permaslug"google/gemini-3-flash-preview-20251217"
provider_name"Google AI Studio"
status200
api_type"completions"
id"gen-1771981281-xW0NnVVNTNSgn6C4mDRZ"
upstream_id(null)
total_cost0.005071
cache_discount(null)
upstream_inference_cost0
provider_name"Google AI Studio"

Evaluation details

Result Evaluator Details Meta Data
100.00% Parse dialogue n/a
dialogue
0
name"Hunter"
dialogue"Excuse me, ma'am? I’m Detective Hunter with the local PD. I’m wondering if I could have a moment of your time regarding the incident near the festival grounds last night."
detectedLang"en"
heavyLang"en"
scores
en0.7944225956381837
tl0.6783216783216783
nl0.5963495963495963
fr0.5876658300466117
no0.5605655330531143
da0.5539177657098526
ro0.5532245532245532
ca0.5484884177463683
it0.523019493985898
pt0.5028102031993084
es0.48314606741573035
sq0.4578029231494578
sv0.4529019980970504
ms0.41416199694345385
de0.413265306122449
cs0.3947368421052632
hr0.38666666666666666
sk0.36323366555924697
sl0.3502824858757062
yo0.3286631640396965
et0.3231312536786345
pl0.3042952208106473
az0.2852703542573026
hu0.26376440460947503
lt0.26187419768934533
is0.20634920634920634
tr0.15191740412979352
fi0.14877868245743894
lv0.14561664190193166
vi0.10366328916601715
be0.05737704918032788
eu0.05737704918032788
uk0.05193734542456719
reliabletrue
passestrue
1
name"Maria"
dialogue"Oh... ¡Hola! Lo siento mucho, pero no entiendo. No hablo nada de inglés. ¿Es usted la policía?"
detectedLang"es"
heavyLang"es"
scores
es0.8153618906942393
pt0.6368390470656595
ca0.6225845410628019
it0.5756958587915818
fr0.5086477987421384
tl0.48218724109362054
no0.421831637372803
en0.4212962962962963
nl0.41369606003752346
ro0.3890518084066471
lt0.3830207305034551
sk0.37810945273631835
hu0.37624750499001997
lv0.35233160621761656
tr0.34963579604578565
yo0.3407172995780591
da0.32578209277238407
eu0.32212581344902386
de0.3169398907103825
cs0.2961711711711712
et0.2799539170506913
sq0.26297169811320753
sv0.24516908212560387
hr0.23218673218673216
fi0.21875
vi0.21875
pl0.20685279187817257
ms0.1883116883116883
az0.18086500655307994
sl0.1270949720670391
reliabletrue
passestrue
2
name"Hunter"
dialogue"I'm sorry, I don't... look, I noticed you were out by the old orchard yesterday with that camera. I need to know if you saw anyone unusual. A man, maybe six-foot-two, wearing a blue flannel?"
detectedLang"en"
heavyLang"en"
scores
en0.8008849557522124
tl0.5964849354375896
da0.39320388349514557
no0.3862520458265139
sv0.3842364532019704
nl0.3704532736429771
fr0.35750999428897773
it0.3428738317757009
cs0.3390129259694477
ms0.33234421364985167
sk0.32594367884961056
pl0.3008079552517091
lv0.29993777224642193
yo0.2857142857142857
de0.2680546519193234
es0.2608409986859396
pt0.25791556728232184
ro0.25595238095238093
ca0.25049966688874087
et0.2474916387959866
hr0.2459785522788204
lt0.23365122615803816
sq0.20156139105748758
hu0.16543026706231453
sl0.15793413173652696
tr0.11556603773584906
is0.10926365795724466
az0.10572337042925278
vi0.10572337042925278
be0.058577405857740586
eu0.058577405857740586
fi0.05303030303030303
uk0.05303030303030303
reliabletrue
passestrue
3
name"Maria"
dialogue"¡Fue horrible! De verdad, estoy muy asustada. Estaba tomando fotos de los manzanos para mi clase cuando vi a alguien entre las sombras. Un hombre... pero luego, simplemente desapareció. Se esfumó en la nada."
detectedLang"es"
heavyLang"es"
scores
es0.804150632215503
pt0.6663544837274643
ca0.6647058823529411
fr0.5694864048338368
it0.5532915360501567
ro0.49324324324324326
tl0.4789762340036563
no0.41018211920529807
en0.40226510067114096
da0.39873417721518983
nl0.3708609271523179
sv0.3672291296625222
tr0.36155913978494625
sq0.3549117247623359
eu0.34240886017535765
et0.3297271872060207
yo0.3181818181818182
de0.31358381502890176
sl0.2892768079800499
lv0.2722165474974464
hr0.26659804426145134
hu0.25974025974025977
sk0.2581988547631442
cs0.2503945291951605
pl0.21789242590559824
ms0.20612813370473537
fi0.18849658314350795
lt0.16812609457092817
az0.16569086651053863
vi0.1512805241215009
is0.12790697674418602
reliabletrue
passestrue
4
name"Hunter"
dialogue"Whoa, slow down. I don't speak Spanish, miss. Did you see a person? A man? 'Hombre'?"
detectedLang"en"
heavyLang"en"
scores
en0.7785651018600531
tl0.6006389776357828
sv0.5348837209302325
da0.5278564683663833
no0.5224450811843362
ca0.48612538540596095
it0.46638207043756674
sq0.4646680942184154
de0.453551912568306
ro0.44008958566629336
cs0.4279176201372997
hu0.4165694282380397
et0.4152046783625731
es0.4033412887828162
nl0.3939393939393939
lv0.3872549019607844
fr0.38650306748466257
lt0.3834771886559803
hr0.38271604938271603
pl0.3670886075949367
pt0.3573264781491003
is0.35233160621761656
sk0.3429697766097241
tr0.299719887955182
yo0.2917847025495751
az0.2774566473988439
ms0.23195084485407066
vi0.21011058451816747
sl0.15254237288135594
be0.12280701754385964
eu0.12280701754385964
fi0.11190053285968028
uk0.11190053285968028
reliabletrue
passestrue
5
name"Maria"
dialogue"¡Sí, un hombre! Pero no fue normal. Estaba allí y un segundo después, el bosque estaba vacío. He intentado mirar las fotos en mi cámara pero tengo miedo de lo que pueda encontrar. Este pueblo es muy extraño, hay algo malo aquí."
detectedLang"es"
heavyLang"es"
scores
es0.8132980004818117
ca0.6808729668519662
pt0.6165264720435428
fr0.5192307692307693
it0.45306986591390264
ro0.4392185238784371
tl0.4153149754809506
en0.3987587276958883
nl0.340144742443593
yo0.3324720068906115
tr0.31957857769973663
de0.31263858093126384
lv0.31263858093126384
sk0.30617726051924804
sv0.30462090623598026
eu0.30430879712746856
hu0.2928832116788321
no0.2824074074074074
sl0.2597898758357211
hr0.2594362159579551
pl0.25837320574162675
da0.25552353506243997
lt0.2253873063468266
ms0.22227797290516807
cs0.21039225674987264
sq0.20918367346938777
vi0.20226453937210498
az0.17200854700854698
is0.1711229946524064
fi0.16666666666666666
et0.11021814006888633
reliabletrue
passestrue
6
name"Hunter"
dialogue"I’m not catching any of that. Look, stay right here. Don't leave the cafe. I’m going to see if Dispatch can find someone to translate, or maybe a teacher from the high school. You clearly saw something that shook you up."
detectedLang"en"
heavyLang"en"
scores
en0.8215488215488216
tl0.6100647439670394
de0.3605212355212355
fr0.3495336278841434
nl0.3440594059405941
no0.3304699343102577
it0.3095362167795727
da0.2981991525423729
cs0.2963356346255975
yo0.29333333333333333
ca0.278322440087146
sv0.26552106430155215
sk0.2647058823529412
ro0.23277359583092067
pt0.23099245502031338
sq0.22785547785547786
es0.22423887587822014
ms0.2219612448620082
hr0.21643997634535778
pl0.18960244648318045
tr0.18209876543209877
et0.18058132343846628
sl0.16927899686520376
lv0.163510101010101
lt0.1603295310519645
az0.13455258001306336
is0.11014103425117529
vi0.10654079568442346
hu0.10533423362592842
eu0.09370725034199727
fi0.045389048991354465
reliabletrue
passestrue
7
name"Maria"
dialogue"Por favor, no me deje sola. No me siento segura. Siento que alguien me está observando desde que salí del festival. ¿Me puede ayudar?"
detectedLang"es"
heavyLang"es"
scores
es0.8059589384076115
pt0.753341820496499
ca0.6552491103202847
it0.6109437751004017
no0.46844993141289437
fr0.4495738636363636
en0.43636363636363634
sq0.4142101284958428
sl0.396417445482866
cs0.37900641025641024
tl0.3683781581092095
ro0.35416666666666663
nl0.34709351305812974
sk0.334192439862543
sv0.3141592920353982
da0.31294326241134746
lt0.30243024302430244
et0.29159049360146255
hr0.28040854224698236
ms0.2533718689788054
pl0.21239837398373984
lv0.2002063983488132
fi0.15115005476451257
az0.13213885778275478
eu0.11123853211009174
tr0.10196987253765932
de0.08284023668639054
hu0.07957244655581948
reliabletrue
passestrue
8
name"Hunter"
dialogue"I have no idea what you're saying, miss. Just... deep breaths. Tranquilo, okay? Is that the word? Just stay put."
detectedLang"en"
heavyLang"en"
scores
en0.809825673534073
tl0.5830437804030577
ca0.4906621392190153
hr0.4163424124513619
da0.4157740993184031
no0.39999999999999997
pt0.3993993993993994
es0.3858751279426817
sv0.3858751279426817
it0.3576017130620985
tr0.3485342019543974
yo0.3435448577680525
nl0.33774834437086093
sq0.3212669683257918
pl0.3111366245694604
lv0.30313588850174217
fr0.2805755395683453
sk0.2736077481840194
et0.27184466019417475
hu0.269183922046285
ms0.2628992628992629
ro0.22380336351875807
cs0.21465968586387435
is0.2052980132450331
vi0.2010652463382157
lt0.1780821917808219
az0.16897506925207756
be0.1044776119402985
de0.10044977511244377
eu0.10044977511244377
uk0.09502262443438914
sl0.08952959028831564
reliabletrue
passestrue
9
name"Hunter"
dialogue"Hunter here. ... What? When? At the old mill? Don't touch anything, I’m five minutes out. Keep the area cordoned off."
detectedLang"en"
heavyLang"en"
scores
en0.808743169398907
tl0.5823389021479713
fr0.381078691423519
da0.36479128856624315
sq0.36363636363636365
no0.360146252285192
ms0.2813141683778234
de0.24812030075187969
sv0.2374727668845316
nl0.23076923076923078
it0.2290748898678414
ca0.22822491730981256
es0.19724770642201836
et0.16963226571767495
sl0.15764139590854392
ro0.15560916767189384
pt0.15254237288135594
yo0.1325898389095415
cs0.0967741935483871
sk0.0967741935483871
fi0.09326424870466321
hr0.08735332464146023
lt0.08256880733944955
pl0.08256880733944955
tr0.077733860342556
reliabletrue
passestrue
10
name"Hunter"
dialogue"Look, I have to go. An emergency call just came in. I’ll come back for you, or I’ll send my partner. Just stay in this cafe, okay? Stay here."
detectedLang"en"
heavyLang"en"
scores
en0.818976279650437
tl0.6470301850048685
da0.5153743315508021
it0.5111260957518543
pl0.4915848527349229
no0.43403590944574555
yo0.4277821625887924
sv0.4052502050861362
cs0.3897306397306397
de0.37928082191780826
ro0.3690165361183638
nl0.3438914027149321
sk0.32745825602968465
sl0.32052483598875353
fr0.31279620853080564
hr0.30754536771728747
ca0.29816069699903197
sq0.2967992240543162
pt0.26321138211382117
lv0.2617107942973524
es0.25488180883864336
lt0.250258531540848
et0.24479166666666666
az0.23844537815126046
ms0.23684210526315788
tr0.2229367631296892
eu0.20242024202420242
vi0.19175027870680048
hu0.15204678362573099
be0.0880503144654088
is0.0799492385786802
uk0.0799492385786802
reliabletrue
passestrue
11
name"Maria"
dialogue"¡No, espere! ¡No se vaya! ¡Todavía no le he enseñado las fotos! ¡Hay algo en el bosque!"
detectedLang"es"
heavyLang"es"
scores
es0.8035714285714286
ca0.6549560853199498
fr0.5719844357976653
pt0.5693030540328896
it0.502262443438914
yo0.4791666666666667
ro0.4696239151398264
en0.46859903381642515
no0.4455645161290322
sl0.4155154091392136
tl0.35672514619883045
hr0.35370152761457113
da0.35217903415783275
sv0.3475682087781732
nl0.3428912783751493
fi0.3349455864570738
tr0.3276283618581907
lv0.3251533742331288
hu0.28848641655886154
cs0.2695883134130146
eu0.2455418381344307
sq0.2413793103448276
sk0.2231638418079096
lt0.20634920634920634
vi0.2040520984081042
de0.19590643274853803
is0.19590643274853803
pl0.19590643274853803
et0.10277324632952692
az0.09688013136288999
reliabletrue
passestrue