NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-06-01-01KT1HAY4C3YBQKVYK4YYVQ4S4
Run Details
Model:
Claude Opus 4.8 (Reasoning, Low)
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
19.7s
Cost:
$0.028440
Provider:
Anthropic
Evaluation summary
Score
Evaluator
Details Link
99.98%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
49.9920%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 500 word scene that has 50% words of dialogue between two thieves named Molly and Bill who are planning to break into a children's party to steal all the presents. Only include the prose of the scene without any markdown or commentary.
Model response
Messages
Meta
AI
Molly spread the blueprints across the kitchen table, knocking over a half-eaten bowl of cereal in the process. "This," she announced, "is the most secure location I've ever seen." Bill leaned over, squinting at the crayon drawing. "Molly, that's a child's drawing of a house. There's a sun with sunglasses in the corner." "Exactly. They've disguised the security blueprints as innocent artwork. Genius." "That's a cat. Or possibly a horse." "Focus, Bill." She tapped the paper with authority. "The Hendersons are throwing little Timmy's seventh birthday party on Saturday. By my estimate, we're looking at fifteen, maybe twenty wrapped presents. Untold riches." Bill frowned. "Untold riches? It's a seven-year-old's birthday. We're talking action figures and maybe a bicycle." "That bicycle could be worth eighty dollars," Molly hissed. "We robbed a jewelry store last month." "And we got caught because you sneezed on the laser grid. This is foolproof. Who suspects two adults at a children's party?" "Everyone, Molly. Everyone suspects two adults at a children's party." She waved him off and pulled a duffel bag onto the table. "That's why we're going undercover. I got us costumes." Bill watched in growing horror as she produced a large, matted clown wig and a unicycle. "Absolutely not." "You're Bobo the Birthday Clown. I'm your assistant, Sparkles." "Why do you get to be Sparkles?" "Because I bought the costumes," she said simply. "Now, the plan. You distract the children with balloon animals while I load the presents into the magic trunk." Bill held up the deflated balloons. "I don't know how to make balloon animals." "Then make balloon... shapes. Children are stupid. They'll clap for anything." "That's a horrible thing to say about children." "We're stealing their birthday presents, Bill. I think we've established we're not good people." He sat down heavily, the wig clutched in his hands. "And how exactly do we transport twenty presents out of a house full of suspicious parents?" Molly grinned and gestured grandly. "The piñata." "The what?" "We bring our own piñata. A massive one. We tell them it's a 'special birthday surprise,' then we stuff the presents inside and carry it out." Bill stared at her for a long moment. "That's actually... not the worst idea you've ever had." "I know." "It's still a terrible idea. The kids will want to hit the piñata." Molly's smile faltered. "We'll tell them it's an antique." "An antique piñata." "A priceless heirloom piñata, Bill. Passed down through generations." He buried his face in his hands. "We're going to prison over a bicycle and some action figures." "We're going to prison in style," she corrected, plopping the clown wig onto his head. "Now practice your balloon shapes, Bobo. The party's in three days, and Sparkles doesn't work with amateurs." Bill looked at his reflection in the toaster and sighed. "I hate everything about this." "You always say that." "And I'm always right," he muttered, twisting a balloon until it popped loudly in his face. Molly nodded approvingly. "The children will love that."
created_at
"2026-06-01T12:08:02.961Z"
6/1/2026, 12:08:02 PM
model
"anthropic/claude-4.8-opus-20260528"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1168
moderation_latency
145
generation_time
19485
tokens_prompt
77
tokens_completion
840
native_tokens_prompt
93
native_tokens_completion
1119
native_tokens_completion_images
(null)
native_tokens_reasoning
0
native_tokens_cached
0
num_media_prompt
(null)
num_input_audio_prompt
(null)
num_media_completion
0
num_search_results
(null)
num_fetches
(null)
web_search_engine
(null)
origin
"https://ncbench.com/"
is_byok
false
finish_reason
"stop"
native_finish_reason
"end_turn"
service_tier
"default"
usage
0.02844
router
(null)
provider_responses
0
endpoint_id
"b0d3beaa-5ff7-4666-a749-b1bb2b14c916"
id
"msg_01GrJDQnMRwUZHGu5tbwLmmp"
is_byok
false
latency
1168
model_permaslug
"anthropic/claude-4.8-opus-20260528"
provider_name
"Anthropic"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.16.0; linux; x64))"
http_referer
(null)
request_id
"req-1780315682-dhdNoXP3NV8vOUPGygkB"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1780315682-pmrIn1cSoVzZjAshBUFJ"
upstream_id
"msg_01GrJDQnMRwUZHGu5tbwLmmp"
total_cost
0.02844
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Anthropic"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
99.98%
Matches word count
n/a
neededClean
false
words
502
0.00%
Dialogue to Total Word Ratio
Ratio: 71.68%, Deviation: 21.68%
neededClean
false
wordsTotal
505
wordsDialogue
362
49.9920%