NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-25-01KYCVYAEG6PXFBVXDC8A4PR6C
Run Details
Model:
Thinking Machines Inkling
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
1m 48s
Cost:
$0.003025
Provider:
DeepInfra
Evaluation summary
Score
Evaluator
Details Link
1.04%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
0.5180%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 500 word scene that has 30% 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
The safe house smelled like instant coffee and poor decisions. Molly pressed her palm against the wall map of Whippersnap Lane, tracing a red circle around 42B. "We are going to a children's party," Bill said. "To steal presents." "Correct." "They are gifts for children. Probably socks." "Wrong. The invitation lists Mr. Pembroke's collection of vintage clockwork automata. He hides them inside candy-striped wrapping paper. We take the boxes, sell the gears, and vanish." Bill chewed his thumbnail. "You want me to wear the clown shoes?" "I want you to wear dignity. But yes, the shoes are in the duffel." Molly unfolded the floor plan. "Entry point is the kitchen window. Mrs. Gable serves punch inside. The living room holds the tree. Thirty-seven parcels, Bill. Thirty-seven." "How do we know the gears are inside?" "Because Pembroke never buys socks. Now stop asking." Bill adjusted his fake mustache. "This feels wrong. Even for us." "Morality is flexible. Your spine should be too." "Last Tuesday you said flexibility caused my hernia." "That was the yoga. Focus." Bill pointed to a yellow sticky note. "What about the dog?" "There is no dog." "There is always a dog." "Not this time. Only a balloon arch and a cake shaped like a dinosaur." "I hate dinosaurs." "Then steal faster." Molly slid the lock picks across the table. "We enter at seven. The parents will be distracted by a magician named Bobo. You distract Bobo. I harvest the tree." "Bobo does card tricks." "Then ask for a heart. Meanwhile I fill the sack." Bill sighed. "What if a child cries?" "Children always cry. We are not the reason." "You stole a teddy bear in March." "That bear was an informant." Bill stood and paced. "I need motivation." "You need a better attitude. The payout buys us winter in Lisbon. Warm sun. No more snow. No more stolen bears." Bill paused. "Lisbon sounds nice." "Exactly. So put on the oversized overalls and practice your balloon animal." "I only know how to make a snake." "Perfect. Snakes scare parents. Parents flee. We win." Bill looked at the clown shoes. "These squeak." "That is the point." Molly zipped her jacket. "We are not villains tonight. We are guests who overstay. We smile, we compliment the cake, we pocket the clockwork, and we disappear before the candles blow out." Bill laced the squeaky shoe. "If I get caught, I am blaming Bobo." "Bobo is imaginary security. Now listen. The sack holds forty pounds max. The automata are heavy brass. Prioritize the square boxes. Ignore the soft ones." "Soft ones are socks." "Exactly." Bill hoisted the duffel. "One question." "Make it quick." "Do we sing Happy Birthday?" Molly grabbed her hat. "Only if they ask nicely. And only if it helps us reach the tree faster." Bill nodded slowly. "I can be festive." "Then let us be festive criminals. Seven sharp. No mistakes. No mercy for the wrapping paper." The door creaked open. Outside, snow fell on Whippersnap Lane like powdered sugar on a very bad idea. Bill followed, squeaking with every step. "This feels like destiny," he said. Molly smiled. "No, Bill. This feels like profit."<|end_message|>
created_at
"2026-07-25T14:47:49.991Z"
7/25/2026, 2:47:49 PM
model
"thinkingmachines/inkling-20260715"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1014
moderation_latency
(null)
generation_time
108000
tokens_prompt
75
tokens_completion
815
native_tokens_prompt
75
native_tokens_completion
735
native_tokens_completion_images
(null)
native_tokens_reasoning
0
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
0.00302519
router
(null)
provider_responses
0
endpoint_id
"d6b9d68e-115a-49d7-81cd-1ad1c6942b0e"
id
"chatcmpl-R7avDlfYgf5WoFLrMOOtkJ98"
is_byok
false
latency
120
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-1784990870-r39COCOed7FPkBdjDH4d"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1784990870-NS1QHU30tB5btvCTqLVd"
upstream_id
"chatcmpl-R7avDlfYgf5WoFLrMOOtkJ98"
total_cost
0.00302519
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
1.04%
Matches word count
n/a
neededClean
false
words
526
0.00%
Dialogue to Total Word Ratio
Ratio: 77.32%, Deviation: 47.32%
neededClean
false
wordsTotal
529
wordsDialogue
409
0.5180%