NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-01-01M1EM2GZ5PGD5E2B0HA85XGR6
Run Details
Model:
Thinking Machines Inkling Small (Reasoning, High)
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
32.4s
Cost:
$0.006245
Provider:
DeepInfra
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
31.75%
Dialogue to Total Word Ratio
(details)
15.8730%
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
The alley smelled like damp pizza and poor life choices. Bill tugged at his black turtleneck and stared at an oversized gift bag he had accidentally purchased at a toy store. "We are breaking into a children's party," Bill said. "Correct," Molly said, tightening her ski mask. "We are stealing presents. This is not a children's party, Bill. It is a vault made of juice boxes and denial." A neon invitation on Bill's windshield promised Magical Fun for All Ages. Molly pointed at it with a gloved finger. "You think they have security?" Bill asked. "I think they have a bouncer named Kevin who knows karate because he watched one documentary," Molly said. Bill nodded. "The cake is basically a pinata of sugar." "Exactly," Molly said. "We slip in during the balloon animal phase, when every adult is blind with wonder and poor judgment." Bill pulled out a map drawn on a napkin. "The presents are in the back room labeled Tresor." "Tresor?" Bill asked. "It is French for treasure," Molly said. "Because the host, a four-year-old named Quentin, is pretentious." Bill sighed. "How do we get past the parents?" "We do not get past them," Molly said. "We become parents. You wear the dad jeans I stole from a lost and found. I wear the mom visor and carry a casserole that is actually a fake." Bill looked at his tactical cargo pants. "These are not dad jeans." "Same difference," Molly said. "You say, Where is the bathroom? and I say, Has anyone seen my pearls? Nobody questions pearls." Bill tapped his earpiece, which was actually a Bluetooth speaker playing lullabies. "The party starts at two. We go at two-fifteen, during the clown's interpretive dance." "Clowns are terrifying," Bill said. "Exactly," Molly said. "He creates a diversion." Bill pulled a present-shaped duffel bag from behind a dumpster. "And then we grab the gifts?" "We grab them," Molly corrected. "But not just any gifts. We grab the expensive electronic ones. The ones that make noise and probably violate at least three safety codes." Bill nodded. "What if there is a guard dog?" "There is a guard dog," Molly said. "It is a golden retriever named Biscuit who only attacks people wearing full black." "Then I wear beige," Bill said. "Too late," Molly said. "You are already wearing black." Bill looked down. "I am wearing black because I am a criminal." "Criminals adapt," Molly said. "You can borrow my mom-jacket. It has a patch that says Team Mom and smells like lavender and regret." Bill put on the jacket. It was enormous and smelled of potpourri. "We need a code word," Bill said. "How about cupcake?" Molly suggested. "Too obvious," Bill said. "Unicorn?" Molly tried. "Also obvious," Bill said. "Glitter bomb?" Molly said. Bill considered it. "Fine. If anything goes wrong, we say glitter bomb and run toward the fence with the loose board next to the trampoline." Bill checked his watch, a cheap plastic one with cartoon dinosaurs. "We should practice our cover story," Bill said. "I will ask where the restroom is," Molly said. "You will ask where the presents are." "I cannot ask where the presents are," Bill protested. "That is suspicious." "You say," Molly instructed, "I heard they have a gift table for the birthday child, and I have a present from a distant relative who is unfortunately named Mr. Gifty." Bill practiced. "Hello, I have a present from Mr. Gifty." "Perfect," Molly said. "You sound like a confused relative." Bill pulled out a set of lockpicks made of spaghetti because they had forgotten real ones. "I have spaghetti picks," Bill said. "Optimistic," Molly said. They walked toward the house, which was decorated with balloons shaped like smiling animals and one ominous pirate skull. "There is the clown," Bill whispered. "The clown is performing an interpretive dance about the ocean," Molly whispered back. "Why is he holding a fish?" Bill asked. "Art," Molly said. They slipped in through the back door, which was held open by a misplaced rubber duck. Inside, the room was chaos: streamers, screaming, and a mountain of wrapped boxes. Bill immediately grabbed a large box labeled LEGOs and hid it under his jacket. "These are heavy," Bill said. "That is good," Molly said. "Heavy means expensive." A child approached them. "Are you here for the party?" the child asked. "We are here for the cupcakes," Molly said quickly. "The cupcakes are in the kitchen," the child said. "Right, but we are also here for the presents," Bill blurted. The child blinked. "You have to sign the guest book." Bill and Molly looked at each other. "We signed it in ink," Molly said. "No," the child said. "You have to write your name and whether you brought a gift." Bill wrote Mr. Gifty and Molly wrote Aunt Glitter. The child nodded and walked away. Bill and Molly moved toward the gift table. "We have three boxes," Molly said. "Only three?" Bill asked. "We need speed," Molly said. "We are not here to build a library." They packed the boxes into the duffel bag. The clown suddenly turned and pointed at them. "Thieves!" the clown shrieked, though his face paint was smeared. "Glitter bomb!" Bill yelled. They ran. Through the kitchen, past the cupcakes, over the trampoline, and through the loose board in the fence. The duffel bag bounced behind them. Molly stopped behind a dumpster. "Did we get them?" Bill unzipped the bag. Inside were three presents, a rubber duck, and a clown nose. "We got them," Bill said. "And we got a clown nose," Molly said. "I will keep the clown nose," Bill said. "It completes my outfit." They walked back to the alley. The alley smelled of damp pizza, regret, and maybe glitter. "Next time," Molly said, "we target a bank." "Next time," Bill said, "I bring spaghetti picks and a better jacket." They split up, leaving behind a trail of balloon animals and the faint scent of lavender.
created_at
"2026-09-01T13:55:53.454Z"
9/1/2026, 1:55:53 PM
model
"thinkingmachines/inkling-small-20260730"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
197
moderation_latency
(null)
generation_time
32121
tokens_prompt
75
tokens_completion
5569
native_tokens_prompt
74
native_tokens_completion
5176
native_tokens_completion_images
(null)
native_tokens_reasoning
4055
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
0.0062445
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788270953-QGUfgyQv3nBJHLYUutpr"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788270953-aRVOyyu8pxIAsS5R7leU"
upstream_id
"chatcmpl-RccNZ32POi5bczNbW78UzYAi"
provider_responses
0
endpoint_id
"d1d9657b-3c3a-49b5-8f81-33204e7c71c3"
id
"chatcmpl-RccNZ32POi5bczNbW78UzYAi"
is_byok
false
latency
77
model_permaslug
"thinkingmachines/inkling-small-20260730"
provider_name
"DeepInfra"
status
200
total_cost
0.0062445
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
997
31.75%
Dialogue to Total Word Ratio
Ratio: 54.79%, Deviation: 4.79%
neededClean
false
wordsTotal
1002
wordsDialogue
549
15.8730%