NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-01-01M1ENT4SA9RSWASGR28TVPJE0
Run Details
Model:
Thinking Machines Inkling Small (Reasoning, High)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
7m 38s
Cost:
$0.012562
Provider:
Together
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
25.19%
Dialogue to Total Word Ratio
(details)
12.5970%
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 garage smelled of old burglar tools and fresh despair, and Molly spread the blueprint of the Henderson residence across the workbench like she was plotting a war rather than a children's party. Bill hovered near the coffee machine, clutching three helium balloons like they were emotional support animals with bad intentions. Afternoon light filtered through the dusty window, illuminating a poster of a cat that said Hang in There, which Molly found deeply ironic. The blueprint showed a suburban living room covered in glitter and regret. There was a bouncy castle drawn in blue ink, a balloon arch sketched in red, and a small dog labeled only Judge. Bill stared at the drawing with the intensity of a man who had just realized he might face a party hat in combat. "The party starts at two," Molly said. "Seventeen minutes." "Seventeen minutes to infiltrate a children's party?" Bill asked. "Exactly," Molly said. "No one suspects thieves at birthdays." "The dog wears a party hat," Bill said. "Judgmental." "We offer a treat," Molly said. "Then we steal." A silence fell over the garage, broken only by the distant sound of a child laughing somewhere in the neighborhood, which felt like a warning. Bill pulled out a notepad with more regret than enthusiasm. "Forty-seven presents," he said. "Some are socks." "We leave the socks," Molly said. "We take the shiny ones." Bill scratched his chin, considering the ethics of sock theft from a four-year-old. "What if they ask who we are?" he asked. Molly smiled with the kind of smile that had launched a thousand bad decisions. "Say you're cousins," she said. "They're four. They won't ask follow-ups." Bill looked at the drawing of the cake, which had candles that looked like they could cause harm. "There's a song," he said. "Do we stop?" "We do not stop for songs," Molly said. "Ambient noise." Bill pointed to a sketch of the snack table. "The parents have cameras," he said. "Hidden ones." "We wear clown noses," Molly said. "It's legal camouflage." Bill pulled out a wig that looked like it had been abandoned in a garage sale years ago. "Should we bring cupcakes?" he asked. "If she takes one," Molly said, "she's distracted." Bill adjusted his backpack, which contained empty gift bags and a crowbar he insisted was a structural tool, not a weapon. "We go in at fourteen past," Molly continued. "We fill bags. We ride the van." "Not a bike?" Bill said. "I always mean the van," Molly said. "But I said bike to confuse the dog." Bill looked through the window at the street, where a stream of small children arrived with balloons and an overwhelming level of optimism. He felt the weight of moral failure settling over him like a wet blanket. "Are we doing this?" he asked. Molly straightened her black turtleneck and gathered her materials. "Bill," she said, "we are thieves. Today the rich are kids. Let's go." Bill followed her into the afternoon sun, wondering if being a villain at a birthday party was any different from being a clown at a bank robbery, and deciding that both probably ended with cake and a mild sense of guilt. They slipped into the neighborhood, ready to steal joy, break gift bags, and become legends among parents who would only remember them as the couple who watered the lawn too loudly. The party was waiting. The presents were vulnerable. And somewhere, a four-year-old was about to have her best birthday, at least until she noticed her gifts were missing.
created_at
"2026-09-01T14:26:15.996Z"
9/1/2026, 2:26:15 PM
model
"thinkingmachines/inkling-small-20260730"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
9001
moderation_latency
(null)
generation_time
457543
tokens_prompt
75
tokens_completion
10721
native_tokens_prompt
75
native_tokens_completion
10437
native_tokens_completion_images
(null)
native_tokens_reasoning
9687
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
0.0125619
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788272776-VgtEmHLTNLBegrsxntTA"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788272776-v9XhTvUR8D3saYjc0H5w"
upstream_id
"68c7bad6-aws_ue1"
provider_responses
0
endpoint_id
"0ed263b7-f746-4eb6-a14c-54f2e1cc2269"
id
"68c7bad6-aws_ue1"
is_byok
false
latency
5740
model_permaslug
"thinkingmachines/inkling-small-20260730"
provider_name
"Together"
status
200
total_cost
0.0125619
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Together"
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
592
25.19%
Dialogue to Total Word Ratio
Ratio: 24.75%, Deviation: 5.25%
neededClean
false
wordsTotal
598
wordsDialogue
148
12.5970%