NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-06-01KK1GW5R5Z4W2JPMVG8B9HDQG
Run Details
Model:
Inception Mercury 2
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
21.9s
Cost:
$0.011897
Provider:
Inception
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
8.63%
Dialogue to Total Word Ratio
(details)
54.3147%
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 old community center smelled of stale pizza and glitter glue, a combination that made Bill’s nose twitch like a cartoon character. Molly perched on a stack of folding chairs, her hair tied in a messy bun that looked like a cotton candy explosion. She tapped a tiny laser pointer against the wall, tracing the outline of the kids’ play tunnel that led straight to the gift table. Bill adjusted his oversized trench coat, which sagged over his lanky frame like a saggy superhero cape, and whispered, “I’ve rigged the karaoke machine to loop ‘Baby Shark’ on a loop. The kids will be too busy doing the fin‑flap dance to notice us.” Molly rolled her eyes, “Fine. I’ll attach a high‑frequency squeak to the clown’s red nose. The guard’s dog will chase it, and the guard will be too busy retrieving the pooch to guard the presents.” The two thieves unfolded a crumpled napkin that served as a blueprint, its doodles of balloons, a towering cake, and a mountain of gifts marked ‘VIP.’ Molly traced a finger along the tunnel entrance, “This is the only way in without stepping on the clown’s juggling pins.” Bill nodded, “Quick, grab the glitter cannon!” Molly replied, “Got it, loading now.” The clown, a lanky man with a painted smile, approached them, “Welcome, dear guests! Who’s ready for a surprise?” Bill leaned in, whispering, “We’re the surprise. Let’s grab the presents and vanish before the cake is cut.” The clown’s nose twitched, emitting a high‑frequency squeak that sent the guard’s dog into a frenzy. The guard shouted, “Come back here, you little monster!” while chasing the dog around the candy table. Molly seized the moment, sprinting to the gift table, her bag already bulging with wrapped treasures. She shouted, “We’ve got the unicorn plush, the LEGO set, the chocolate fountain—everything!” Bill joined her, “And the best part is we’ll be the only ones who get to open them later.” As they turned to leave, a child shouted, “Where’s the cake?” Bill answered, “It’s on its way, just minute!” The kids cheered, the clown laughed, and the thieves slipped out, pockets full and conscience empty. “This is insane!” Molly muttered. “Let’s make it quick.” “No alarms, just candy.” “We’re in the clear.” After disappearing into night, Molly and Bill slipped into shadows behind community center, echo of distant laughter fading behind them. The stolen loot rustled in bags as they hurried toward alley, moonlight glinting off glitter that clung to shoes. They paused under flickering streetlamp, exchanging a satisfied grin that said more than any word could. “We should sell the LEGO set online,” Molly suggested, but Bill waved a hand, remembering police’s AI‑driven surveillance drones that patrolled downtown streets. Instead, they agreed to dump loot at charity warehouse, where items would be redistributed to children in need—a attempt at redemption that felt more like a joke than a conscience. As light painted sky pink, thieves vanished into city’s maze, plotting next caper.
created_at
"2026-03-06T12:10:06.475Z"
3/6/2026, 12:10:06 PM
model
"inception/mercury-2-20260304"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
21469
moderation_latency
(null)
generation_time
21859
tokens_prompt
77
tokens_completion
6641
native_tokens_prompt
60
native_tokens_completion
15844
native_tokens_completion_images
(null)
native_tokens_reasoning
15181
native_tokens_cached
5
num_media_prompt
(null)
num_input_audio_prompt
(null)
num_media_completion
0
num_search_results
(null)
origin
"https://ncbench.com/"
is_byok
false
finish_reason
"stop"
native_finish_reason
"stop"
usage
0.011896875
router
(null)
provider_responses
0
endpoint_id
"c2a24e34-c306-4b4d-9e2f-b162c2d8bf0e"
id
"chatcmpl-b1a8f0cb-b94f-464c-b242-4cd191870832"
is_byok
false
latency
105
model_permaslug
"inception/mercury-2-20260304"
provider_name
"Inception"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.14.0; linux; x64))"
http_referer
(null)
api_type
"completions"
id
"gen-1772799006-RVaLFb7rIDpLE39p2pPQ"
upstream_id
"chatcmpl-b1a8f0cb-b94f-464c-b242-4cd191870832"
total_cost
0.011896875
cache_discount
0.000001125
upstream_inference_cost
0
provider_name
"Inception"
Evaluation details
Result
Evaluator
Details
Meta Data
100.00%
Matches word count
n/a
neededClean
false
words
500
8.63%
Dialogue to Total Word Ratio
Ratio: 37.00%, Deviation: 7.00%
neededClean
false
wordsTotal
500
wordsDialogue
185
54.3147%