Run: 2026-05-27-01KSMF9Y49F0WQXAWA77SRJ1XD

Run Details

Model:
Skyfall 36B V2
Test:
Language Writing
Scenario:
Character dialogue (French) in a story
Duration:
7.4s
Cost:
$0.000433
Provider:
Parasail

Evaluation summary

Score Evaluator
Details Link
71.43% 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="Marie"> Marie is a 30-year old college student on a study abroad program. Born in France, she is fluent in French 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 French, 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: Marie 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 French, and Hunter looks confused. They talk for a while, until Hunter gets a call and has to leave. Return the dialogue between Marie and Hunter as a voice sheet: Marie: "..." Hunter: "..."

Model response

AI
After logging in the details about characters and places, it's time to create a voice sheet for a scene between Marie and Hunter. MARIE (looking troubled, shaking): "<sigh> Ah non, c'était terrible ce soir-là. J'ai regardé par la fenêtre et j'ai vu... oh non, je ne peux pas le dire." MARIE (raising her voice, words jumbled): "<Frant> I tried to run but they came at me. One man, Missin'... Guzmán something. Then I saw Hunter, but he didn't see me! Mon dieu, it was terrible!" HUNTER (frowning, confused): "She doesn't speak English? Dammit, I should've brought one of our translators... Ah, shit." (Marriesa): "No way. musical.Private." HUNTER ( Getting closer to mime what she's saying, talking louder): "THEY HERE? Bookstore." MARIE (snapping, shaking her head rapidly): "NO! No way! It's not- listen- just..." (She looks pained, looking away.) HUNTER (sighing sadly, looking down in defeat): "Thank you, Marie. Did,. Here, take this." (Taking out a card and writing a number on the back.) HUNTER (Stuffing the card in her hand, nodding): "Call me if you need. anytime. I'm gonna get to the bottom,no matter what the lieutenant says." (He nods in sincerity as he gets up and leaves.) (Marie whispers to herself, before taking a deep breath and looking after him, shaking her head. The tension in the air is overwhelming, the weight of a hundred stories floating between every gesture and phrase exchanged.)
created_at"2026-05-27T10:22:25.431Z" 5/27/2026, 10:22:25 AM
model"thedrummer/skyfall-36b-v2"
app_id182717
external_user(null)
streamedtrue
cancelledfalse
latency214
moderation_latency(null)
generation_time7276
tokens_prompt316
tokens_completion373
native_tokens_prompt269
native_tokens_completion368
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)
usage0.00043275
router(null)
provider_responses
0
endpoint_id"1eb01ded-ae11-49e6-8aa6-3067584070bd"
id"chatcmpl-b464f70d1b3b90ad"
is_byokfalse
latency214
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-1779877345-MAUApZdTJjK3y8QAPv1d"
session_id(null)
api_type"completions"
id"gen-1779877345-cjDgnxulMLh0KMqZSr0z"
upstream_id"chatcmpl-b464f70d1b3b90ad"
total_cost0.00043275
cache_discount0.0000096
upstream_inference_cost0
provider_name"Parasail"
response_cache_source_id(null)

Evaluation details

Result Evaluator Details Meta Data
71.43% Parse dialogue n/a
dialogue
0
name"MARIE"
dialogue"<sigh> Ah non, c'était terrible ce soir-là. J'ai regardé par la fenêtre et j'ai vu... oh non, je ne peux pas le dire."
detectedLang"fr"
heavyLang"fr"
scores
fr0.7932175351530191
it0.5435179549604382
ca0.47442186405045556
ro0.44029850746268656
es0.4186046511627907
yo0.38271604938271603
sl0.3578767123287671
nl0.3528904227782571
sq0.3397887323943662
en0.3194192377495462
da0.2877492877492877
hr0.27466150870406186
no0.2732558139534884
lv0.25595238095238093
hu0.24471299093655588
vi0.24318869828456105
tl0.2378048780487805
cs0.23391215526046988
et0.22680412371134018
lt0.21218487394957983
sv0.17218543046357618
pl0.16666666666666666
fi0.1638795986622074
sk0.16201117318435754
de0.1601343784994401
tr0.11137440758293839
eu0.1039426523297491
ms0.10071942446043164
pt0.08759124087591241
az0.07749077490774908
reliabletrue
passestrue
1
name"MARIE"
dialogue"<Frant> I tried to run but they came at me. One man, Missin'... Guzmán something. Then I saw Hunter, but he didn't see me! Mon dieu, it was terrible!"
detectedLang"en"
heavyLang"en"
scores
en0.7911832946635731
tl0.5712243925678895
fr0.5210218201170835
da0.5194874532835024
no0.4943820224719101
ca0.47674418604651164
nl0.43431803896920174
yo0.4260204081632653
es0.4039735099337748
de0.4027869940278699
sq0.3885869565217392
sv0.3839835728952772
it0.3482983345401882
pl0.3464052287581699
et0.336283185840708
lv0.3273542600896861
pt0.3263473053892215
hr0.3176648976497347
fi0.3028659953524399
ms0.296875
sk0.29467084639498436
sl0.27594529364440873
hu0.2653061224489796
cs0.2537313432835821
ro0.24369747899159663
lt0.2248062015503876
az0.22010398613518195
is0.1818181818181818
tr0.1758241758241758
eu0.0816326530612245
hi0.07975460122699386
be0.07216494845360825
uk0.06542056074766354
vi0.06542056074766354
reliabletrue
passesfalse
2
name"HUNTER"
dialogue"She doesn't speak English? Dammit, I should've brought one of our translators... Ah, shit."
detectedLang"en"
heavyLang"en"
scores
en0.7873094647288196
tl0.49409780775716694
hr0.4163424124513619
ro0.4106090373280943
sv0.39332659251769464
no0.3902439024390244
ca0.3877551020408163
pl0.3736951983298539
lv0.3288590604026846
sq0.318955732122588
da0.3135011441647597
it0.3095512082853855
de0.27007299270072993
yo0.25
lt0.24717691342534504
nl0.23954372623574144
cs0.21465968586387435
es0.20212765957446807
is0.20212765957446807
vi0.2010652463382157
sk0.1836734693877551
tr0.18144611186903137
ms0.1792065663474692
sl0.17355371900826447
az0.16897506925207756
et0.16897506925207756
fr0.11764705882352942
pt0.11764705882352942
be0.1044776119402985
eu0.10044977511244377
hu0.10044977511244377
uk0.09502262443438914
reliabletrue
passestrue
3
name"HUNTER"
dialogue"THEY HERE? Bookstore."
detectedLang"en"
heavyLang"en"
scores
en0.7967479674796748
lv0.6212121212121212
sq0.6044303797468354
it0.48559670781893005
no0.4541484716157206
da0.44933920704845814
eu0.4075829383886255
sk0.3902439024390244
hr0.36224489795918363
reliabletrue
passestrue
4
name"MARIE"
dialogue"NO! No way! It's not- listen- just..."
detectedLang"en"
heavyLang"en"
scores
en0.7979214780600462
no0.6111111111111112
sv0.6040723981900453
tr0.571078431372549
lv0.5535714285714286
tl0.5535714285714286
ca0.5478036175710594
pt0.5345744680851064
es0.5333333333333334
fi0.5179063360881543
de0.515235457063712
hr0.5014245014245015
da0.49567723342939485
sk0.4615384615384615
pl0.4479495268138801
it0.44267515923566875
eu0.4006849315068493
nl0.3986254295532647
et0.36823104693140796
hu0.3137254901960784
is0.2857142857142857
cs0.2647058823529412
fr0.2647058823529412
ro0.2647058823529412
sq0.2647058823529412
sl0.25213675213675213
reliabletrue
passesfalse
5
name"HUNTER"
dialogue"Thank you, Marie. Did,. Here, take this."
detectedLang"en"
heavyLang"en"
scores
en0.8106060606060606
tl0.5967741935483871
ro0.44306930693069313
it0.4201030927835051
sq0.4201030927835051
sl0.39999999999999997
fr0.39189189189189183
no0.39189189189189183
vi0.39189189189189183
da0.3869209809264305
sv0.3869209809264305
cs0.3784530386740331
hr0.25
pt0.24242424242424243
ca0.23728813559322032
es0.22945205479452058
lt0.22945205479452058
pl0.22945205479452058
de0.21875
et0.21875
eu0.21875
ms0.21875
nl0.21875
sk0.21875
yo0.21875
fi0.2077464788732394
lv0.2077464788732394
reliabletrue
passestrue
6
name"HUNTER"
dialogue"Call me if you need. anytime. I'm gonna get to the bottom,no matter what the lieutenant says."
detectedLang"en"
heavyLang"en"
scores
en0.7945694891032512
no0.5110544217687075
it0.5047372954349698
fr0.4893428063943161
tl0.4666048237476809
sv0.4346116027531956
et0.38105489773950485
pt0.3653421633554084
sq0.3539325842696629
da0.33983926521239954
fi0.32273262661955243
nl0.31137724550898205
lv0.3055555555555556
yo0.29792429792429787
es0.2936117936117936
ca0.28393524283935245
sk0.27944862155388467
hr0.27398989898989895
pl0.267515923566879
cs0.26658163265306123
sl0.26658163265306123
lt0.2493472584856397
de0.22506738544474394
ms0.21981004070556312
tr0.21981004070556312
ro0.21555252387448842
is0.14814814814814814
reliabletrue
passestrue