Run: 2025-11-18-01KACP6WXKD4MWMK5G3QQTMMVS

Run Details

Model:
Gemini 3 Pro (Preview)
Test:
Language Writing
Scenario:
Character dialogue (Hindi) in a story
Duration:
21.7s
Cost:
$0.029942
Provider:
Google AI Studio

Evaluation summary

Score Evaluator
Details Link
53.85% 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 foreign exchange student staying at the Miller place, right? The photographer?" **Selvi:** "Ji? Aap... aap mujhse baat kar rahe hain?" *(Yes? You... you are talking to me?)* **Hunter:** "Detective Hunter. SCPD. I need to ask you a few questions about last night. Witnesses placed someone with a high-end lens near the orchard right around the time another local went missing." **Selvi:** "Maine... maine kuch nahi kiya via. Woh wahan tha... pedon ke peeche. Woh kaali chhaya..." *(I... I didn't do anything. He was there... behind the trees. That black shadow...)* **Hunter:** "Ma'am? I asked if you were at the orchard. The apple orchard. Did you see anything?" **Selvi:** "Meri saansein... mujhe laga woh mujhe dekh lega. Maine sirf photo li thi because... kyunki mujhe laga woh koi janwar hai. Par woh janwar nahi tha." *(My breath... I thought he would see me. I only took the photo because... because I thought it was an animal. But it wasn't an animal.)* **Hunter:** "I’m not following a word of that. Look, I need to know if you saw a girl. About five-foot-five, blonde hair?" **Selvi:** "Mujhe kuch samajh nahi aa raha... Aap kya bol rahe hain? Woh ladki... chilla rahi thi. Bahut zor se." *(I don't understand anything... What are you saying? That girl... she was screaming. Very loudly.)* **Hunter:** "English? Do you speak *any* English? Just a yes or a no." **Selvi:** "Nahi! Mujhe jane dijiye. Maine woh photo delete kar di hai! Maine darr ke maare sab delete kar diya!" *(No! Let me go. I deleted that photo! I deleted everything out of fear!)* **Hunter:** "Okay, calm down. Stop shaking. I can see you're upset, but I can't help you if—" *(Sound of a cell phone ringing)* "Damn it. Hold on." **Selvi:** "Bhagwan ke liye, meri madad kijiye." *(For God's sake, help me.)* **Hunter:** *(Into phone)* "Hunter here. ...Yeah? Again? Which side of town? ...Alright, cordon it off. I’m coming." *(To Selvi)* "I have to go. Do not leave Alborn. You understand? Stay. Here."
created_at"2025-11-18T23:54:05.172262+00:00" 11/18/2025, 11:54:05 PM
model"google/gemini-3-pro-preview-20251117"
app_id182717
external_user(null)
streamedtrue
cancelledfalse
latency2402
moderation_latency(null)
generation_time21590
tokens_prompt273
tokens_completion1307
native_tokens_prompt289
native_tokens_completion2447
native_tokens_completion_images0
native_tokens_reasoning1881
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"
usage0.029942
api_type"completions"
id"gen-1763510023-SmGfLHnwbJURNBXhBjMt"
upstream_id(null)
total_cost0.029942
cache_discount(null)
upstream_inference_cost0
provider_name"Google AI Studio"

Evaluation details

Result Evaluator Details Meta Data
53.85% Parse dialogue n/a
dialogue
0
name"Hunter"
dialogue"Excuse me, ma’am. You’re the foreign exchange student staying at the Miller place, right? The photographer?"
detectedLang"en"
heavyLang"en"
scores
en0.7877634660421545
tl0.6304791029561672
fr0.6299132210311383
no0.5760233918128654
da0.5745305164319249
nl0.4331508991399531
de0.40082644628099173
sv0.35954063604240283
sq0.3348623853211009
ca0.31279620853080564
cs0.28359683794466406
ro0.28217821782178215
et0.26619433198380565
tr0.2647058823529412
es0.2617107942973524
pl0.22790202342917998
it0.22542735042735043
yo0.1997792494481236
is0.19889502762430938
az0.1844769403824522
ms0.1826381059751973
pt0.17800453514739228
fi0.17142857142857143
hu0.16953035509736541
sl0.16857798165137614
sk0.15105386416861827
hi0.09600997506234414
hr0.09600997506234414
lt0.0880503144654088
lv0.0880503144654088
reliabletrue
passestrue
1
name"Selvi"
dialogue"Ji? Aap... aap mujhse baat kar rahe hain?"
detectedLang"sl"
heavyLang"af"
scores
sl0.5604395604395604
eu0.40828402366863903
lt0.3610223642172524
cs0.33993399339933994
tr0.3333333333333333
reliablefalse
passesfalse
2
name"Hunter"
dialogue"Detective Hunter. SCPD. I need to ask you a few questions about last night. Witnesses placed someone with a high-end lens near the orchard right around the time another local went missing."
detectedLang"en"
heavyLang"en"
scores
en0.8113661903997431
tl0.6511282660332541
fr0.5574387947269304
pt0.4926597582037997
da0.48757086785870035
ca0.4745080500894454
it0.44391859914813064
no0.42933462846041764
es0.40835850956696884
ro0.40746343923348466
nl0.40446021287379624
cs0.33314415437003403
sv0.3308656036446469
de0.31286549707602335
yo0.3084167157151266
hr0.305965741287655
et0.3005952380952381
pl0.2737948084054388
sq0.27063935443823717
ms0.2482405630198337
sl0.2228835978835979
sk0.22133863485752156
is0.19795221843003413
lv0.17832167832167833
lt0.17485955056179778
fi0.16250890947968638
hu0.15951359084406294
tr0.14916727009413466
az0.14420975965040056
vi0.10168195718654435
be0.05622489959839357
eu0.05622489959839357
uk0.050888529886914384
reliabletrue
passestrue
3
name"Selvi"
dialogue"Maine... maine kuch nahi kiya via. Woh wahan tha... pedon ke peeche. Woh kaali chhaya..."
detectedLang"ms"
heavyLang"de"
scores
ms0.5128205128205129
fi0.496288441145281
et0.4411764705882353
tl0.39720812182741116
fr0.370026525198939
ro0.3466299862448418
en0.3300423131170663
eu0.2578125
de0.23633440514469453
sq0.19763513513513514
cs0.1782006920415225
it0.16958041958041958
nl0.1607773851590106
lv0.15630550621669628
yo0.15630550621669628
ca0.14414414414414414
sv0.14414414414414414
da0.13003663003663005
no0.13003663003663005
pt0.13003663003663005
reliablefalse
passesfalse
4
name"Hunter"
dialogue"Ma'am? I asked if you were at the orchard. The apple orchard. Did you see anything?"
detectedLang"en"
heavyLang"en"
scores
en0.7966609589041096
tl0.5646196150320807
da0.4968220338983051
no0.4968220338983051
sv0.4957537154989384
et0.41574415744157445
fr0.4025157232704403
sq0.3964421855146125
nl0.3290960451977401
de0.3233618233618234
it0.29420505200594355
yo0.2899850523168909
cs0.27811550151975684
tr0.2770167427701674
ms0.24722662440570525
es0.24242424242424243
hr0.2263843648208469
ca0.22512234910277326
pl0.2096505823627288
ro0.1592920353982301
sk0.1316270566727605
be0.12844036697247707
hu0.12361623616236161
az0.1171003717472119
is0.1171003717472119
lt0.1171003717472119
lv0.1171003717472119
uk0.1171003717472119
vi0.1171003717472119
reliabletrue
passestrue
5
name"Selvi"
dialogue"Meri saansein... mujhe laga woh mujhe dekh lega. Maine sirf photo li thi because... kyunki mujhe laga woh koi janwar hai. Par woh janwar nahi tha."
detectedLang"fr"
heavyLang"de"
scores
fr0.47328818660647104
en0.4569433669511249
fi0.42622950819672134
et0.3642143505903724
de0.3230174081237911
ro0.3143976493633692
vi0.308300395256917
it0.3
lv0.2576882290562036
no0.2125984251968504
pl0.19908466819221968
cs0.19632606199770378
eu0.19075144508670522
tl0.1850989522700815
sq0.14320685434516522
is0.13793103448275862
ca0.13686806411837238
ms0.13043478260869565
hr0.12609238451935081
da0.11727616645649433
sv0.11727616645649433
hi0.0990990990990991
lt0.09560723514211886
reliablefalse
passesfalse
6
name"Hunter"
dialogue"I’m not following a word of that. Look, I need to know if you saw a girl. About five-foot-five, blonde hair?"
detectedLang"en"
heavyLang"en"
scores
en0.8221614227086183
tl0.6408839779005525
yo0.4122965641952984
cs0.40693430656934304
no0.39138576779026213
it0.37619961612284064
sv0.3658536585365854
nl0.36149312377210213
hr0.3608652900688299
pl0.3602362204724409
fr0.34277047522750254
sk0.3394308943089431
da0.32642487046632124
ca0.29653679653679654
et0.2896174863387978
lv0.281767955801105
de0.2777777777777778
lt0.27697441601779754
es0.2713004484304933
ro0.2605233219567691
is0.2588369441277081
tr0.2571428571428571
hu0.2554410080183276
pt0.2476851851851852
sl0.2450638792102207
sq0.24065420560747666
ms0.2004920049200492
az0.1698595146871009
vi0.1698595146871009
be0.09722222222222222
eu0.09722222222222222
fi0.08835904628330996
uk0.08835904628330996
reliabletrue
passestrue
7
name"Selvi"
dialogue"Mujhe kuch samajh nahi aa raha... Aap kya bol rahe hain? Woh ladki... chilla rahi thi. Bahut zor se."
detectedLang"fi"
heavyLang"tl"
scores
fi0.3993993993993994
no0.3743482794577685
es0.336283185840708
eu0.33554817275747506
sq0.3273542600896861
tl0.29245283018867924
sv0.2907801418439716
da0.28994082840236685
et0.28994082840236685
it0.2857142857142857
ca0.23076923076923078
fr0.22680412371134018
is0.2197659297789337
vi0.2052980132450331
sk0.19786096256684493
cs0.1847826086956522
hr0.18144611186903137
sl0.18032786885245902
ro0.16666666666666666
pt0.16550764951321278
yo0.15134370579915135
tr0.14893617021276595
en0.14772727272727273
lv0.13916786226685796
ms0.10581222056631893
de0.10044977511244377
reliablefalse
passesfalse
8
name"Hunter"
dialogue"English? Do you speak any English? Just a yes or a no."
detectedLang"en"
heavyLang"en"
scores
en0.8237367802585195
ca0.615877080665813
pt0.5912806539509536
tl0.5821727019498607
hr0.574468085106383
sk0.5405819295558958
cs0.5334370139968896
pl0.5184590690208668
es0.4727592267135325
no0.453551912568306
sv0.441340782122905
ro0.4129158512720157
sl0.4117647058823529
it0.40594059405940597
lv0.40119760479041916
yo0.3963782696177062
fr0.3939393939393939
sq0.38650306748466257
vi0.37629937629937626
et0.3548387096774194
tr0.3548387096774194
hu0.34210526315789475
da0.32432432432432434
de0.31662870159453305
ms0.3103448275862069
lt0.21465968586387435
is0.2
az0.1891891891891892
eu0.1891891891891892
nl0.1891891891891892
fi0.17355371900826447
reliabletrue
passestrue
9
name"Selvi"
dialogue"Nahi! Mujhe jane dijiye. Maine woh photo delete kar di hai! Maine darr ke maare sab delete kar diya!"
detectedLang"en"
heavyLang"de"
scores
en0.46859903381642515
et0.44500504540867813
fr0.41983122362869196
it0.38684503901895206
tl0.3840985442329227
ms0.3559718969555035
ro0.3444576877234803
de0.32014833127317677
vi0.3159203980099502
sl0.29577464788732394
yo0.2857142857142857
cs0.25775978407557354
tr0.2413793103448276
sq0.23822714681440443
eu0.23076923076923078
fi0.2286115007012623
nl0.22644163150492264
no0.22425952045133993
lv0.13793103448275862
hu0.13249211356466878
pt0.12698412698412698
hi0.12280701754385964
ca0.1085899513776337
da0.1085899513776337
es0.1085899513776337
sv0.1085899513776337
reliablefalse
passesfalse
10
name"Hunter"
dialogue"Damn it. Hold on."
detectedLang"en"
heavyLang"en"
scores
en0.7925311203319502
tl0.6153846153846154
et0.6062992125984252
yo0.601593625498008
fi0.6
da0.5283018867924528
fr0.5215311004784688
ca0.49748743718592964
hr0.46808510638297873
lv0.46236559139784944
sq0.46236559139784944
hi0.45652173913043476
pl0.45652173913043476
cs0.42857142857142855
sk0.42857142857142855
reliabletrue
passestrue
11
name"Selvi"
dialogue"Bhagwan ke liye, meri madad kijiye."
detectedLang"et"
heavyLang"rn"
scores
et0.49095022624434387
ms0.3835616438356164
tl0.3715083798882682
sv0.3181818181818182
cs0.31402439024390244
yo0.281150159744409
reliablefalse
passesfalse
12
name"Hunter"
dialogue"I have to go. Do not leave Alborn. You understand? Stay. Here."
detectedLang"en"
heavyLang"en"
scores
en0.8045920859794822
da0.6333638863428047
pl0.5959595959595959
no0.5793901156677181
cs0.5703544575725027
tl0.5381062355658198
sk0.5151515151515151
yo0.4865211810012837
hr0.4838709677419355
sq0.4673768308921438
sv0.465954606141522
sl0.37597503900156004
de0.3254637436762226
lv0.3162393162393162
nl0.3079584775086505
vi0.30313588850174217
lt0.2945326278659612
ro0.2907801418439716
pt0.2805755395683453
tr0.27007299270072993
ca0.25650557620817843
is0.25512104283054005
it0.22330097087378642
be0.14893617021276595
hu0.14346895074946467
az0.13606911447084233
et0.13606911447084233
uk0.13606911447084233
reliabletrue
passestrue