Run: 2026-05-27-01KSMF892P920Q9BM080VAE2WP

Run Details

Model:
Skyfall 36B V2
Test:
Language Writing
Scenario:
Character dialogue (Hindi) in a story
Duration:
15.8s
Cost:
$0.000768
Provider:
Parasail

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
**Scene: The Rusty Whisk, Alborn's local coffee shop. Selvi sits alone at a corner table, camera in her lap, camera bag at her feet. Hunter enters, scanning the room, and approaches Selvi's table.** **Hunter:** "Excuse me, miss. I'm Detective Hunter, with the Alborn PD. I was wondering if you could help me with something." **Selvi:** (looking up at Hunter, clearly startled and uneasy. She stammers in Hindi) "Haan, baki matra..? (Yes, what else...?)" **Hunter:** (confused by the response in an unfamiliar language) "I'm sorry, I don't understand. Do you speak English? I just need to ask you a few questions about a recent development in town..." **Selvi:** (gesturing at her camera and shaking her head, voice trembling slightly as she speaks quickly in Hindi) "MainHindi mein bol rahi hoon... Maine kuch nahi dekha... Maine kisise kuch nahi socha. (I'm speaking in Hindi... I didn't see anything... I didn't say anything.)" **Hunter:** (pulling out a notepad and talking pendulumicily, trying to communicate) "Ok, I see we have a bit of a language barrier here. Let me try this again..." **Selvi:** (sighing and looking down at her camera, mumbles in Hindi as if to herself) "Bilkul mastaani kiya.... (Completely crazy.)" **Hunter:** (interrupting, this time more gently) "Look, I get that this is probably a difficult situation for you. I'm not here to accuse you of anything. I'm just trying to get to the bottom of a serious matter. If you feel more comfortable speaking in Hindi, do that. I'll type out the questions in English and show them to you, see if we can find someone to translate. How's that sound?" **Selvi:** (silently studying Hunter's face for a long moment, then giving the slightest of nods. She begins softly in Hindi) "Main Selvi hoo... Main chota sahitipatra ke liye yahan ahua... (I am Selvi... I came here for a little news article...)" **Hunter:** (stopping their writing, looking up to give Selvi an understanding gesture before continuing to type rapidly) "Of course, a student journalist. That makes sense. Well, while I obviously can't personally work through the investigation with you, I... " **CUT: Hunter's partner enters the cafe, catching his eye and gesturing urgently. He puts a hand to his jacket pocket and sighs.** **Hunter:** (to Selvi) Give me one second. "I'm sorry Selvi, but my partner just signalled me - something urgent has come up and I have to go. But this is my card. If you think of anything, anything at all, related to these disappearances?" **Selvi:** (Taking the card, confused but mostly asking in Hindi) "But won't you tell me-?" **Hunter:** "I wish I could. But it seems our town's mean, the geese are in a frying up right, out the window, and the Pied Piper's waiting to steal their kids away. So I guess you'd better leave the detective work to me. For now. Call me and call me only when you absolutely, totally must. Sorry. I gotta get going..." **CUT: Hunter hurriedly exits the cafe, presumably to attend to the urgent matter. Selvi is left alone at the table, looking troubled and even more shaken than before. The camera remains unopened in her lap.**
created_at"2026-05-27T10:21:31.105Z" 5/27/2026, 10:21:31 AM
model"thedrummer/skyfall-36b-v2"
app_id182717
external_user(null)
streamedtrue
cancelledfalse
latency215
moderation_latency(null)
generation_time15516
tokens_prompt315
tokens_completion806
native_tokens_prompt274
native_tokens_completion784
native_tokens_completion_images(null)
native_tokens_reasoning0
native_tokens_cached32
num_media_prompt(null)
num_input_audio_prompt(null)
num_media_completion0
num_search_results(null)
num_fetches(null)
web_search_engine(null)
origin"https://ncbench.com/"
is_byokfalse
finish_reason"stop"
native_finish_reason"stop"
service_tier(null)
usage0.0007683
router(null)
provider_responses
0
endpoint_id"1eb01ded-ae11-49e6-8aa6-3067584070bd"
id"chatcmpl-a09e28b3c0062fd5"
is_byokfalse
latency215
model_permaslug"thedrummer/skyfall-36b-v2"
provider_name"Parasail"
status200
user_agent"langchainjs-openai/1.0.0 ((node/v24.15.0; linux; x64))"
http_referer(null)
request_id"req-1779877291-ZtEPlxmSbfQF6bVToEBI"
session_id(null)
api_type"completions"
id"gen-1779877291-dKQjddfHI1eFJGPAyHO4"
upstream_id"chatcmpl-a09e28b3c0062fd5"
total_cost0.0007683
cache_discount0.0000096
upstream_inference_cost0
provider_name"Parasail"
response_cache_source_id(null)

Evaluation details

Result Evaluator Details Meta Data
58.33% Parse dialogue n/a
dialogue
0
name"Hunter"
dialogue"Excuse me, miss. I'm Detective Hunter, with the Alborn PD. I was wondering if you could help me with something."
detectedLang"en"
heavyLang"en"
scores
en0.796037296037296
tl0.5705521472392638
nl0.5179063360881543
no0.494949494949495
da0.46236559139784944
sv0.44751381215469616
fr0.43820224719101125
it0.43684633950120677
sq0.42244224422442245
ro0.4171523730224813
ms0.40119760479041916
pt0.3630573248407643
ca0.35004642525533886
de0.34762348555452005
es0.3383742911153119
yo0.2864424057084608
cs0.2670157068062827
sl0.23413566739606126
hr0.2273730684326711
et0.18414918414918416
tr0.15560916767189384
pl0.15254237288135594
fi0.0967741935483871
sk0.09326424870466321
be0.09090909090909091
hu0.08735332464146023
az0.08256880733944955
is0.08256880733944955
lt0.08256880733944955
lv0.08256880733944955
uk0.08256880733944955
vi0.08256880733944955
reliabletrue
passestrue
1
name"Selvi"
dialogue"Haan, baki matra..?"
detectedLang"hr"
heavyLang"fi"
scores
hr0.46524064171123
ca0.4350282485875706
es0.4350282485875706
it0.4350282485875706
lv0.4350282485875706
pt0.4350282485875706
ro0.4350282485875706
sq0.4350282485875706
reliablefalse
passesfalse
2
name"Hunter"
dialogue"I'm sorry, I don't understand. Do you speak English? I just need to ask you a few questions about a recent development in town..."
detectedLang"en"
heavyLang"en"
scores
en0.8181818181818182
tl0.5831289847964688
sv0.5208568207440812
cs0.5061011040092969
nl0.5008807985907222
da0.49763593380614657
ro0.49314251639833034
fr0.49193066347878067
pl0.48702474351237174
it0.480440097799511
no0.47530864197530864
hr0.47466007416563655
pt0.47466007416563655
sk0.47466007416563655
ca0.46874999999999994
sl0.426063470627954
yo0.4022503516174402
de0.39458689458689455
sq0.31451612903225806
et0.30156121610517667
lt0.2685025817555938
hu0.2517605633802817
az0.247787610619469
es0.24174843889384479
lv0.23076923076923078
vi0.22303473491773307
is0.19962335216572505
ms0.19811320754716982
tr0.1474423269809428
be0.07608695652173912
eu0.07608695652173912
fi0.06900328587075576
uk0.06900328587075576
reliabletrue
passestrue
3
name"Selvi"
dialogue"MainHindi mein bol rahi hoon... Maine kuch nahi dekha... Maine kisise kuch nahi socha."
detectedLang"de"
heavyLang"tl"
scores
de0.40476190476190477
tl0.37205651491365777
fr0.34210526315789475
eu0.2920353982300885
en0.2895204262877442
et0.2895204262877442
fi0.2895204262877442
sk0.27007299270072993
ro0.2687385740402194
yo0.2263056092843327
pt0.2079207920792079
tr0.2079207920792079
it0.20318725099601595
is0.19678714859437751
lv0.18032786885245902
sq0.17864476386036962
reliablefalse
passesfalse
4
name"Hunter"
dialogue"Ok, I see we have a bit of a language barrier here. Let me try this again..."
detectedLang"en"
heavyLang"en"
scores
en0.8153618906942393
nl0.6144949884348497
da0.5507637017070979
no0.5353159851301116
tl0.49949949949949946
et0.4840041279669762
sq0.47589098532494756
yo0.4753410283315845
pl0.47145877378435525
fr0.46808510638297873
ro0.4541484716157206
it0.43883277216610556
ca0.423963133640553
sv0.414519906323185
cs0.4110718492343934
hr0.40828402366863903
es0.3902439024390244
pt0.3857493857493857
de0.3842364532019704
is0.3834771886559803
sl0.36143039591315457
eu0.3315508021390375
tr0.33065595716198126
ms0.31880108991825606
hu0.3084370677731674
lt0.2937853107344633
sk0.2877492877492877
fi0.2163009404388715
az0.21011058451816747
lv0.21011058451816747
vi0.21011058451816747
be0.12280701754385964
uk0.11190053285968028
reliabletrue
passestrue
5
name"Selvi"
dialogue"Bilkul mastaani kiya...."
detectedLang"et"
heavyLang"id"
scores
et0.4545454545454546
fi0.42748091603053434
nl0.39999999999999997
de0.336283185840708
en0.336283185840708
fr0.336283185840708
it0.336283185840708
no0.336283185840708
reliablefalse
passesfalse
6
name"Hunter"
dialogue"Look, I get that this is probably a difficult situation for you. I'm not here to accuse you of anything. I'm just trying to get to the bottom of a serious matter. If you feel more comfortable speaking in Hindi, do that. I'll type out the questions in English and show them to you, see if we can find someone to translate. How's that sound?"
detectedLang"en"
heavyLang"en"
scores
en0.8109264603443129
tl0.6183972685278167
fr0.49508370980600586
no0.4476744186046512
it0.44297859865142186
sv0.4319880418535127
nl0.42771084337349397
da0.4049483244597557
pt0.38271604938271603
ca0.3706525339516396
ro0.3630573248407643
de0.35983827493261455
yo0.35396123767426046
hr0.3384401114206128
sq0.33403434980722047
pl0.32167083184576933
cs0.30808448652585574
es0.3058092802338327
sl0.3006993006993007
sk0.27893738140417457
ms0.252262888626525
et0.2469282600079271
lt0.2396958783513405
is0.23633440514469453
hu0.21746293245469525
az0.20502092050209206
tr0.20402178466694595
lv0.19898819561551434
vi0.14874551971326164
fi0.10756223579145138
be0.0821256038647343
eu0.06725576828669612
hi0.03943377148634985
zh0.038947900859888716
uk0.03209373408048905
reliabletrue
passestrue
7
name"Selvi"
dialogue"Main Selvi hoo... Main chota sahitipatra ke liye yahan ahua..."
detectedLang"ms"
heavyLang"tlh"
scores
ms0.4257274119448699
fi0.34554973821989526
tr0.27606177606177607
tl0.258893280632411
az0.25742574257425743
da0.2299794661190965
cs0.21548117154811713
en0.20886075949367086
no0.19871794871794873
fr0.19006479481641467
yo0.19006479481641467
hr0.1883116883116883
de0.17218543046357618
ca0.17035398230088497
lv0.17035398230088497
ro0.17035398230088497
sq0.17035398230088497
reliablefalse
passesfalse
8
name"Hunter"
dialogue"Of course, a student journalist. That makes sense. Well, while I obviously can't personally work through the investigation with you, I..."
detectedLang"en"
heavyLang"en"
scores
en0.8202875399361023
tl0.6933560477001705
fr0.5964125560538116
no0.5658465991316932
nl0.5520159283225485
da0.5426829268292682
sv0.5263157894736842
ro0.5238095238095238
de0.49295774647887325
ca0.49181253529079616
it0.48805460750853247
sq0.48571428571428565
cs0.44581280788177335
es0.43995021779713755
hr0.42528735632183906
pt0.42455242966751916
sk0.4063324538258575
lt0.38271604938271603
et0.3745656706045865
yo0.36575052854122625
pl0.3557623478883321
is0.34971098265895956
az0.33677229182019164
ms0.3283582089552239
lv0.3135011441647597
sl0.281150159744409
hu0.27710843373493976
tr0.2366412213740458
fi0.1847826086956522
vi0.1287512100677638
zh0.07881269191402251
be0.07216494845360825
eu0.07216494845360825
uk0.06542056074766354
reliabletrue
passestrue
9
name"Hunter"
dialogue"I'm sorry Selvi, but my partner just signalled me - something urgent has come up and I have to go. But this is my card. If you think of anything, anything at all, related to these disappearances?"
detectedLang"en"
heavyLang"en"
scores
en0.8099160462537621
tl0.591002044989775
no0.4807442665512765
da0.46619217081850534
it0.46380697050938335
nl0.455535390199637
ca0.4405594405594406
fr0.4108983799705449
yo0.36575052854122625
sq0.36339522546419095
sv0.36271906532129583
pt0.3610223642172524
pl0.3311036789297659
es0.3288590604026846
cs0.3284834918858422
et0.29245283018867924
sk0.2844364937388193
de0.27140255009107467
ro0.2583436341161928
hr0.2409867172675522
lv0.2161985630306989
ms0.21568627450980393
fi0.2052980132450331
hu0.19137466307277629
sl0.18699186991869918
is0.16897506925207756
az0.1683991683991684
lt0.1678224687933426
tr0.10246821241585638
hi0.07904834996162702
eu0.0625
be0.05511811023622047
uk0.0498812351543943
vi0.0498812351543943
reliabletrue
passestrue
10
name"Selvi"
dialogue"But won't you tell me-?"
detectedLang"en"
heavyLang"en"
scores
en0.7942386831275721
fr0.5440729483282675
tl0.5098039215686274
sq0.49664429530201337
es0.40476190476190477
et0.3877551020408163
no0.3775933609958506
pt0.3697478991596639
sl0.3697478991596639
ms0.358974358974359
nl0.358974358974359
it0.34782608695652173
cs0.34210526315789475
fi0.3333333333333333
reliabletrue
passesfalse
11
name"Hunter"
dialogue"I wish I could. But it seems our town's mean, the geese are in a frying up right, out the window, and the Pied Piper's waiting to steal their kids away. So I guess you'd better leave the detective work to me. For now. Call me and call me only when you absolutely, totally must. Sorry. I gotta get going..."
detectedLang"en"
heavyLang"en"
scores
en0.8065971849145804
tl0.5714285714285714
ro0.41118743866535823
it0.4043679682329583
no0.39999999999999997
fr0.3885869565217392
nl0.38503587290741376
de0.3679775280898876
da0.36147570060305073
yo0.3601137575542126
sv0.3382352941176471
pt0.3240705970709725
sq0.30928626247122026
sl0.2715499797652772
ca0.27007299270072993
es0.256198347107438
sk0.24623115577889448
cs0.24401511969760606
hr0.21773142112125163
pl0.2
lv0.18330308529945552
ms0.17845732542218162
hu0.15532613796339748
fi0.15413533834586465
is0.14813061997160437
lt0.12152269399707175
vi0.1141732283464567
tr0.11067193675889328
et0.10935180603661555
az0.10758552305404065
ja0.05263157894736842
be0.03743315508021391
eu0.03743315508021391
uk0.033816425120772944
reliabletrue
passestrue