Run: 2026-02-08-01KGYZVHMF8MZHF85S7MBKB2EV

Run Details

Model:
Claude Haiku 4.5
Test:
Language Writing
Scenario:
Character dialogue (Hindi) in a story
Duration:
6.9s
Cost:
$0.003322
Provider:
Amazon Bedrock

Evaluation summary

Score Evaluator
Details Link
60.00% 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="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
# Voice Sheet: Cafe Encounter **Setting:** *Alborn Cafe. Selvi sits at a corner table, scrolling through her camera's photos. Hunter approaches with his notepad.* --- Selvi: *(looking up, startled)* "Haan? Kya hua?" Hunter: "Miss, do you have a minute? I'm Detective Hunter. I'm investigating some disappearances around town. Were you near the old Miller property last night?" Selvi: *(visibly shaken, speaking rapidly in Hindi)* "Nahi, nahi... Main wahan nahi tha. Lekin maine kuch ajeeb dekha tha. Bahut daraavna tha." Hunter: *(frowning, tilting his head)* "I'm sorry, I don't... can you speak English?" Selvi: *(frustrated, gesturing)* "English? Nahi, nahi... Hindi. Sirf Hindi bolti hoon main." Hunter: *(sighing, pulling up his phone's translation app)* "Okay, give me a second here. What did you see?" Selvi: *(still trembling)* "Raat ko maine lights dekhi thin. Ajeeb lights. Aur ek ladka tha... ya ladki tha... samajh nahi aya. Bahut tez se cycle le kar chala gaya." Hunter: *(reading the translation on his phone screen)* "Lights? And someone on a bike?" Selvi: *(nodding desperately)* "Haan, haan! Cycle! Main photograph lena chahti tha, lekin... lekin mare hath kaamp rahe the." Hunter: "You tried to photograph them?" Selvi: *(shaking her head)* "Nahi... dar lag gaya. Mujhe malum nahi tha kya karna chahiye." Hunter: *(leaning forward)* "Can you show me what time this was on your camera's timestamp? Maybe—" *(Hunter's phone buzzes loudly. He glances at it, his expression hardening.)* Hunter: "I have to take this. Don't leave town, okay? We may need to talk again." Selvi: *(quietly, in Hindi)* "Theek hai. Main yahi hoon." Hunter: *(already turning away, speaking into his phone)* "Yeah, this is Hunter. What do we have?" *(He walks out of the cafe, leaving Selvi alone with her camera, still visibly shaken.)*
created_at"2026-02-08T16:02:54.865705+00:00" 2/8/2026, 4:02:54 PM
model"anthropic/claude-4.5-haiku-20251001"
app_id182717
external_user(null)
streamedtrue
cancelledfalse
latency656
moderation_latency152
generation_time6693
tokens_prompt262
tokens_completion498
native_tokens_prompt307
native_tokens_completion603
native_tokens_completion_images(null)
native_tokens_reasoning0
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"
usage0.003322
router(null)
provider_responses
0
id"896ee4d4-13f6-4646-8f3a-f951fce0b9e3"
status200
is_byokfalse
latency656
endpoint_id"d000261c-5aeb-47d1-ac87-ad64dbe1cdb6"
provider_name"Amazon Bedrock"
model_permaslug"anthropic/claude-4.5-haiku-20251001"
api_type"completions"
id"gen-1770566567-MAhmusK9fFz5FgBmjsU1"
upstream_id"896ee4d4-13f6-4646-8f3a-f951fce0b9e3"
total_cost0.003322
cache_discount(null)
upstream_inference_cost0
provider_name"Amazon Bedrock"

Evaluation details

Result Evaluator Details Meta Data
60.00% Parse dialogue n/a
dialogue
0
name"Selvi"
dialogue"Haan? Kya hua?"
detectedLang""
heavyLang"fi"
scores
reliablefalse
passestrue
1
name"Hunter"
dialogue"Miss, do you have a minute? I'm Detective Hunter. I'm investigating some disappearances around town. Were you near the old Miller property last night?"
detectedLang"en"
heavyLang"en"
scores
en0.8101898101898102
tl0.6489283074648927
fr0.5941905168731312
pt0.5437079731027858
ro0.5388349514563107
da0.5297029702970297
it0.5194739504299444
ca0.5153061224489796
no0.4986807387862797
nl0.4739756367663345
de0.47045707915273133
es0.45181765724177725
sv0.43653618030842234
hr0.4290865384615385
cs0.42073170731707316
sl0.4124922696351268
sq0.4110353378797272
et0.40176322418136023
ms0.38351719662556777
sk0.35983827493261455
pl0.2846385542168675
yo0.2846385542168675
az0.2731446059678654
lt0.2647058823529412
tr0.24242424242424243
is0.24060751398880897
lv0.23633440514469453
fi0.22067268252666117
hu0.20502092050209206
vi0.16003536693191867
eu0.13003663003663005
reliabletrue
passestrue
2
name"Selvi"
dialogue"Nahi, nahi... Main wahan nahi tha. Lekin maine kuch ajeeb dekha tha. Bahut daraavna tha."
detectedLang"eu"
heavyLang"et"
scores
eu0.4558991981672394
fi0.42073170731707316
ms0.4092039800995025
de0.40624999999999994
tl0.3863049095607235
fr0.38391699092088194
en0.35549525101763907
sq0.3185078909612626
et0.2554858934169279
ro0.23633440514469453
hr0.1908006814310051
sl0.1908006814310051
is0.15630550621669628
lv0.15630550621669628
yo0.13793103448275862
reliablefalse
passesfalse
3
name"Hunter"
dialogue"I'm sorry, I don't... can you speak English?"
detectedLang"en"
heavyLang"en"
scores
en0.8045602605863192
tl0.5405819295558958
no0.4307400379506641
ca0.4174757281553398
it0.4174757281553398
da0.4129158512720157
hr0.4129158512720157
sv0.4117647058823529
cs0.37888198757763975
ro0.35064935064935066
tr0.3288590604026846
az0.3181818181818182
sq0.3087557603686636
pl0.29577464788732394
pt0.25
yo0.2248062015503876
de0.1935483870967742
ms0.1935483870967742
sk0.1935483870967742
be0.1891891891891892
hu0.18256130790190736
et0.17355371900826447
is0.17355371900826447
lt0.17355371900826447
lv0.17355371900826447
uk0.17355371900826447
vi0.17355371900826447
reliabletrue
passestrue
4
name"Selvi"
dialogue"English? Nahi, nahi... Hindi. Sirf Hindi bolti hoon main."
detectedLang"en"
heavyLang"tl"
scores
en0.5551601423487544
tl0.5543672014260249
is0.5479204339963833
it0.47589098532494756
eu0.39759036144578314
de0.375
et0.2732558139534884
fr0.26035502958579887
sq0.258160237388724
no0.24698795180722888
hi0.2378048780487805
da0.22360248447204967
hr0.22360248447204967
ms0.22360248447204967
ro0.22360248447204967
sv0.22360248447204967
reliablefalse
passesfalse
5
name"Hunter"
dialogue"Okay, give me a second here. What did you see?"
detectedLang"en"
heavyLang"en"
scores
en0.8176392572944298
tl0.6082621082621082
fr0.5782208588957055
et0.5439469320066335
it0.5378151260504201
sq0.5378151260504201
es0.45972495088408644
nl0.45436507936507936
pt0.4532803180914513
ro0.4376278118609407
cs0.43298969072164945
da0.41737288135593226
sl0.39293598233995586
no0.3692660550458715
hu0.36194895591647336
ms0.3483412322274882
fi0.3341404358353511
ca0.3309002433090025
sk0.32098765432098764
yo0.30203045685279184
hr0.289405684754522
pl0.2763157894736842
lt0.22969187675070027
tr0.21875
is0.21428571428571427
az0.20289855072463767
eu0.20289855072463767
lv0.20289855072463767
vi0.20289855072463767
de0.19590643274853803
sv0.1863905325443787
reliabletrue
passestrue
6
name"Selvi"
dialogue"Raat ko maine lights dekhi thin. Ajeeb lights. Aur ek ladka tha... ya ladki tha... samajh nahi aya. Bahut tez se cycle le kar chala gaya."
detectedLang"tl"
heavyLang"en"
scores
tl0.5183887915936953
en0.4687701223438506
fr0.45364238410596025
sl0.39159292035398224
ro0.3707093821510297
yo0.35090479937057434
tr0.336283185840708
es0.30904522613065327
fi0.30904522613065327
eu0.30203045685279184
sq0.2881794650560828
lv0.27184466019417475
et0.2445054945054945
it0.22680412371134018
az0.2112810707456979
cs0.20520231213872833
da0.17746759720837485
no0.175
ms0.1716867469879518
sv0.16750756811301715
de0.15123456790123457
hr0.13883089770354906
pt0.1260593220338983
is0.11003236245954692
nl0.1081081081081081
hu0.10617551462621885
ca0.10423452768729642
lt0.10326086956521739
sk0.08536585365853659
reliablefalse
passesfalse
7
name"Hunter"
dialogue"Lights? And someone on a bike?"
detectedLang"en"
heavyLang"en"
scores
en0.8188405797101449
tl0.7001499250374813
yo0.5943204868154158
ca0.5402298850574713
fr0.5305164319248826
pt0.5215311004784688
et0.5203836930455635
fi0.5157384987893463
es0.5121951219512195
cs0.5085995085995086
sk0.5061728395061729
hr0.5
pl0.48586118251928023
hu0.43820224719101125
ro0.41002949852507375
it0.38461538461538464
sl0.3103448275862069
hi0.29577464788732394
lt0.2907801418439716
tr0.2779783393501805
is0.2727272727272727
da0.2647058823529412
sq0.2647058823529412
az0.25925925925925924
eu0.25925925925925924
lv0.25925925925925924
nl0.25925925925925924
no0.25925925925925924
vi0.25925925925925924
de0.250936329588015
ms0.23954372623574144
sv0.23954372623574144
reliabletrue
passestrue
8
name"Selvi"
dialogue"Haan, haan! Cycle! Main photograph lena chahti tha, lekin... lekin mare hath kaamp rahe the."
detectedLang"en"
heavyLang"tlh"
scores
en0.6218152866242038
fr0.496288441145281
tl0.3716931216931217
fi0.302496328928047
sq0.291044776119403
sv0.27811550151975684
eu0.23633440514469453
ro0.19491525423728812
et0.1908006814310051
yo0.18384879725085912
no0.16666666666666666
is0.15630550621669628
nl0.15026833631484796
ca0.1410488245931284
da0.1410488245931284
de0.1410488245931284
it0.1410488245931284
ms0.1410488245931284
hi0.1394927536231884
reliabletrue
passesfalse
9
name"Hunter"
dialogue"You tried to photograph them?"
detectedLang"en"
heavyLang"en"
scores
en0.8296422487223168
tl0.5485327313769752
sk0.5370370370370371
cs0.511002444987775
fr0.5024875621890548
yo0.4065281899109792
pl0.40298507462686567
lv0.37888198757763975
sl0.37694704049844235
hr0.3670886075949367
no0.35275080906148865
da0.34210526315789475
lt0.34210526315789475
hi0.2779783393501805
reliabletrue
passestrue
10
name"Selvi"
dialogue"Nahi... dar lag gaya. Mujhe malum nahi tha kya karna chahiye."
detectedLang"tl"
heavyLang"tl"
scores
tl0.46236559139784944
sv0.3902439024390244
ro0.36363636363636365
ms0.34210526315789475
no0.3269230769230769
eu0.32038834951456313
nl0.32038834951456313
lt0.2647058823529412
pl0.2505353319057816
sq0.2505353319057816
is0.22048997772828507
es0.20993227990970656
pt0.20993227990970656
de0.18032786885245902
sl0.17061611374407584
reliablefalse
passesfalse
11
name"Hunter"
dialogue"Can you show me what time this was on your camera's timestamp? Maybe—"
detectedLang"en"
heavyLang"en"
scores
en0.788135593220339
tl0.5754716981132075
nl0.49324324324324326
et0.42528735632183906
no0.4155844155844156
fi0.3935309973045822
fr0.38016528925619836
ca0.3580599144079886
pt0.3562231759656652
it0.3487698986975398
hr0.34687953555878087
cs0.3283582089552239
de0.29245283018867924
es0.2730210016155089
da0.27184466019417475
ms0.2647058823529412
yo0.2647058823529412
pl0.26108374384236455
sq0.2474916387959866
sl0.16356877323420074
ro0.16044776119402984
hi0.15730337078651685
az0.14611005692599618
tr0.14611005692599618
sk0.14285714285714285
lt0.12959381044487428
reliabletrue
passestrue
12
name"Hunter"
dialogue"I have to take this. Don't leave town, okay? We may need to talk again."
detectedLang"en"
heavyLang"en"
scores
en0.8262469337694195
tl0.6181491464510332
da0.4785276073619632
pl0.4613434727503169
nl0.4370860927152318
yo0.4310575635876841
hr0.375
no0.37407952871870404
cs0.35703479576399394
sk0.32432432432432434
et0.32216905901116427
sl0.3189102564102564
lv0.30327868852459017
lt0.28209459459459457
vi0.2685025817555938
ca0.24511545293072823
sv0.24511545293072823
sq0.23971377459749554
it0.21296296296296297
ro0.17475728155339806
be0.1414141414141414
tr0.1414141414141414
hu0.1361788617886179
az0.1290983606557377
is0.1290983606557377
uk0.1290983606557377
reliabletrue
passestrue
13
name"Selvi"
dialogue"Theek hai. Main yahi hoon."
detectedLang"vi"
heavyLang"nl"
scores
vi0.511400651465798
nl0.4117647058823529
it0.39999999999999997
en0.39759036144578314
fr0.3697478991596639
de0.34210526315789475
reliablefalse
passesfalse
14
name"Hunter"
dialogue"Yeah, this is Hunter. What do we have?"
detectedLang"en"
heavyLang"en"
scores
en0.8266563944530047
nl0.6100519930675911
pl0.5222929936305732
da0.5044052863436124
pt0.49324324324324326
yo0.46808510638297873
sl0.44306930693069313
sq0.43890274314214467
tl0.4260204081632653
hu0.4063324538258575
cs0.3715083798882682
sk0.3697478991596639
hr0.34782608695652173
vi0.33035714285714285
is0.2902208201892744
de0.281150159744409
no0.25742574257425743
fr0.24242424242424243
sv0.23728813559322032
it0.21875
ca0.2077464788732394
ms0.2077464788732394
ro0.2077464788732394
tr0.2077464788732394
reliabletrue
passestrue