NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-01-01M1ENEHSN439Z4QQC8T0QT85M
Run Details
Model:
Thinking Machines Inkling Small (Reasoning, High)
Test:
Dialogue tags
Scenario:
Write 500 words with 70% dialogue
Duration:
1m 10s
Cost:
$0.004670
Provider:
DeepInfra
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
3.92%
Dialogue to Total Word Ratio
(details)
1.9588%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 500 word scene that has 70% 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 leaned against the stolen laundry van and checked her watch, the words Fresh Sheets glinting in the late afternoon sun. Bill stood beside her wearing a tuxedo jacket over a Hawaiian shirt, a false mustache, and an eye patch he insisted made him look like a pirate who had given up on life. Molly sighed. “Bill, the invitation said bring a gift, not a criminal record.” “Criminal records are endorsements of hustle,” Bill said, adjusting his beret, which had previously belonged to a silent mime. “Plus I have this eye patch. It makes me mysterious.” “Mysterious, yes. Capable of breaking into a children’s birthday party, debatable.” “We’re not breaking in,” Bill corrected. “We’re liberating presents from a room full of juice boxes and broken dreams.” Molly checked her clipboard. “There’s a clown.” “The clown is the key,” Bill said. He pulled out a napkin map. “Look. The clown is crying into a balloon animal near the gift table because he lost his shoe in the pool. His emotional state renders him non-functional for security purposes.” “He’s method acting.” “He’s weeping, Molly. No security professional can operate while simultaneously sobbing into latex.” Molly glanced at the yard where balloons swayed and children screamed near a bounce house that smelled of rubber and regret. “We need to get past the parents. There’s a patio door.” “The patio door is unlocked because the father is drunk on sparkling cider and the mother is obsessed with gluten-free cupcakes,” Bill said. “We’re walking into a fortress of denial.” “We’re walking in because you wear a mustache that says Uncle Greg.” “It’s strategic,” Bill said. “Uncle Greg doesn’t belong at this party, which is exactly why he blends in. He just stands near the gift table and looks concerned.” “I’ll create a diversion,” Molly said. “I’ll pretend to be a lost parent named Greg.” “You already are Greg,” Bill pointed out. “It’s layered.” “Here’s the plan,” Molly said, holding up a plastic unicorn flashlight. “You go through the patio, grab the presents near the bounce house, and don’t talk. You talk too much.” “I talk when there’s air in my lungs,” Bill said. “It’s a medical condition.” “The gift table is unguarded because parents are too busy arguing about whether a twelve-year-old needs a gift card for a video game,” Molly said. “The presents are huge. Wrapped in paper that cost more than this van, which is literally a laundry truck with Fresh Sheets written on the side.” “We provide a service,” Bill said proudly. “We steal, but we smell clean.” They slipped through the fence as a black and white clown balloon arch swayed above them. A child pointed at Bill. “Are you the clown’s assistant?” Bill froze, then nodded. “Yes. I’m Single-Parent Consultant.” “Why are you here?” “Because I forgot the rope in my other pants,” Bill said smoothly. Molly grabbed Bill’s elbow. “We have a window. The piñata is coming down in sixty seconds. I’ll knock it over to create chaos. You grab the largest box.” “Which box?” “The one that says To Emma, Age Six, From Everyone. That’s the motherlode.” “Got it,” Bill said. Molly approached a cluster of parents arguing over cupcake frosting. “Excuse me, I’m looking for my son. He has a peanut allergy and a mustache.” A mother turned. “You don’t look like a parent.” “I am the ghost of parenting past,” Molly said. She knocked into a table, sending a piñata swinging wildly toward a group of screaming children. “Run!” she whispered. Bill darted to the gift table, stuffed two large wrapped boxes under his arms, and bolted toward the patio. Molly followed, clutching her clipboard. They reached the fence and slammed into the van. “We did it,” Molly said, breathing hard. “We stole from children,” Bill said, holding up the boxes. “We redistributed wealth,” Molly corrected. “Also, I think I left the mustache in the yard.” “We can buy another,” Bill said. “We have the sheets to sell.” Bill started the engine. “Next time we hit an adult party. They have wine and less judgment.” “We need a better disguise for that,” Molly said. “We have the van,” Bill said. “That’s enough.” They drove away, the stolen presents piled beside them like golden monuments to poor choices, leaving behind only a strawberry-scented mustache and a very confused clown.
created_at
"2026-09-01T14:19:56.092Z"
9/1/2026, 2:19:56 PM
model
"thinkingmachines/inkling-small-20260730"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
201
moderation_latency
(null)
generation_time
60332
tokens_prompt
75
tokens_completion
4596
native_tokens_prompt
74
native_tokens_completion
3864
native_tokens_completion_images
(null)
native_tokens_reasoning
3401
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
"stop"
service_tier
(null)
usage
0.0046701
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788272396-Gy8WubECsfp39IOVmPwc"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788272396-iUSjBo3Bh3zQ2cNoez0c"
upstream_id
"chatcmpl-REGhCMCAgwpQfStqrffstXdb"
provider_responses
0
endpoint_id
"0ed263b7-f746-4eb6-a14c-54f2e1cc2269"
is_byok
false
latency
10077
model_permaslug
"thinkingmachines/inkling-small-20260730"
provider_name
"Together"
status
504
1
endpoint_id
"d1d9657b-3c3a-49b5-8f81-33204e7c71c3"
id
"chatcmpl-REGhCMCAgwpQfStqrffstXdb"
is_byok
false
latency
81
model_permaslug
"thinkingmachines/inkling-small-20260730"
provider_name
"DeepInfra"
status
200
total_cost
0.0046701
cache_discount
(null)
upstream_inference_cost
0
provider_name
"DeepInfra"
response_cache_source_id
(null)
data_region
"global"
workspace_id
"97e315e5-d303-487d-83c1-83180e8a13d4"
Evaluation details
Result
Evaluator
Details
Meta Data
0.00%
Matches word count
n/a
neededClean
false
words
722
3.92%
Dialogue to Total Word Ratio
Ratio: 61.95%, Deviation: 8.05%
neededClean
false
wordsTotal
728
wordsDialogue
451
1.9588%