NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-25-01KYCTH4H4G57RP59D9PEWHKK6
Run Details
Model:
Thinking Machines Inkling (Reasoning)
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
14m 22s
Cost:
$0.043622
Provider:
DeepInfra
Evaluation summary
Score
Evaluator
Details Link
86.38%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
43.1902%
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
Bill flattened the blueprint across the sticky booth. Molly stabbed her spoon into a milkshake that had achieved sidewalk consistency. "We're doing it," Bill said. "Saturday. Three p.m. Henderson twins' party at the community center. Forty-seven registered gifts." Molly didn't look up. "Forty-seven? That's not a heist. That's a logistics nightmare with juice boxes." Bill tapped the map. "Exactly. We need camouflage. No ski masks." "So we dress as clowns?" Molly asked. "I have boundaries." "Not clowns. Caterers." Bill straightened an imaginary tie. "Black pants, trays of shrimp puffs. We glide in like culinary ghosts." "You want to serve appetizers while robbing toddlers?" Molly snorted. "Bill, that's not misdirection, that's catering with a felony." "Psychology," Bill said. "You feed them canapés, they lower their guard. I bag the loot." Molly stirred her shake. "What loot? Lego and stuffed animals?" "High-value items," Bill said. "Gaming console. Maybe a drone." "It's a five-year-old's birthday," Molly said. "The only high-value item is the sugar crash." Bill traced a line on the map. "Entry through the kitchen service door. Latch older than democracy. My cousin works maintenance. He owes me from the raccoon wedding." "I don't want details," Molly said. "Good. Exit through the balloon arch. We collapse it behind us—chaos. Load the van, painted like a party truck." "Balloons in February?" Molly said. "Nobody buys balloons in winter." "They will when ours floats away carrying forty-seven presents," Bill said. Molly sipped. "What about the piñata?" "We leave it," Bill said. "It's a donkey full of candy and broken dreams," Molly said. "I'm taking it." "Fine. But we leave the cake. I'm not a monster." "You're stealing gifts from children," Molly said. "The monster train already left the station." Bill folded his hands. "We are redistributors of joy. Those kids have too many toys. We have rent." "That sounds like a villain monologue," Molly said. "In this economy," Bill said, "everyone's a villain." Molly pushed her cup away. "What's my job?" Bill slid over a name tag: MOLLY—EVENT SPECIALIST. "Event specialist?" "You smile, hand out napkins, ask about allergies," Bill said. "Meanwhile I wear an apron that says 'Chef Bill' and fill black garbage bags." "Garbage bags? For stolen birthday presents?" "Classy black ones," Bill said. "What if we're caught?" "Use your charm," Bill said. "Remember the security guard?" "He was lonely and possibly hallucinating," Molly said. "But okay. I'll tell the twins I'm a fairy godmother with a tax problem." "Perfect. Meanwhile I avoid Mr. Waffles." "There's a dog?" "Small. Don't make eye contact." Molly stood, pulling on her worn coat. "This is your worst plan. And you've proposed a submarine heist." Bill stood, leaving a tip of loose change. "This is art, Molly. High art with shrimp puffs." Molly headed for the door. "If I'm arrested, I'm blaming you." Bill raised his coffee cup. "To the heist." Molly pulled her collar up. "To prison. At least we'll have appetizers."
created_at
"2026-07-25T14:23:09.356Z"
7/25/2026, 2:23:09 PM
model
"thinkingmachines/inkling-20260715"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1428
moderation_latency
(null)
generation_time
861498
tokens_prompt
75
tokens_completion
9599
native_tokens_prompt
75
native_tokens_completion
10759
native_tokens_completion_images
(null)
native_tokens_reasoning
8805
native_tokens_cached
32
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
"stop"
service_tier
(null)
usage
0.04362239
router
(null)
provider_responses
0
endpoint_id
"d6b9d68e-115a-49d7-81cd-1ad1c6942b0e"
id
"chatcmpl-RPL76rYZ3rxcwBLtvo1tLCbB"
is_byok
false
latency
62
model_permaslug
"thinkingmachines/inkling-20260715"
provider_name
"DeepInfra"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.18.0; linux; x64))"
http_referer
(null)
request_id
"req-1784989389-ycgrJFzGvGoyl2g8pIW3"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1784989389-ZwT6EkN5fuC1clllUxu8"
upstream_id
"chatcmpl-RPL76rYZ3rxcwBLtvo1tLCbB"
total_cost
0.04362239
cache_discount
0.00002656
upstream_inference_cost
0
provider_name
"DeepInfra"
response_cache_source_id
(null)
data_region
"global"
Evaluation details
Result
Evaluator
Details
Meta Data
86.38%
Matches word count
n/a
neededClean
false
words
489
0.00%
Dialogue to Total Word Ratio
Ratio: 73.79%, Deviation: 23.79%
neededClean
false
wordsTotal
496
wordsDialogue
366
43.1902%