NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-10-01M25DF7831DSF8KFGXEXAQX3P
Run Details
Model:
DeepSeek V4.1 Flash (Reasoning, High)
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
2m 57s
Cost:
$0.011645
Provider:
DeepInfra
Evaluation summary
Score
Evaluator
Details Link
99.98%
Matches word count
(details)
99.95%
Dialogue to Total Word Ratio
(details)
99.9673%
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
Molly spread the birthday invitation across the sticky diner table, smoothing its crayon unicorn with her thumb. The party would start at two. Forty children, one magician, a piñata, and a mountain of presents. Bill stirred his coffee and frowned at the frosting stains on the map she had drawn. "We hit the community center at one forty-five," Molly said. "Caterers. Fake cake. We wheel the presents out under the tray." Bill squinted. "The cake is hollow?" "The cake is a lie," Molly said. "The tray is reinforced." He tapped a ketchup stain that represented the bounce house. "What about the kids?" "Piñata. Blindfolds. Screaming. We have six minutes." "Last time you said six minutes, I lived in a ball pit for three hours." "You looked happy." "I was hiding from a toddler with a foam sword." Molly ignored him and traced the service corridor with a fry. "You take the gifts. I handle the crowd. Cash envelopes go in the diaper bag." "We don't have a baby." "We have a diaper bag. Close enough." Bill rubbed his temples. "And if someone asks why two grown adults are carrying a giant fake cake and a diaper bag?" "We tell them we're the entertainment." "Another clown?" "Backup clown. Clowns are unpredictable, but so are children. It's a fair fight." Bill leaned back, staring at the ceiling. The plan had holes. The magician might notice a tricycle vanishing. The birthday girl might notice her pile shrinking. The grandma with the camera might post everything to Facebook before they reached the parking lot. He said as much, slowly, while Molly drew arrows around the kitchen exit. "We give the birthday girl a decoy pony," she said. "A piñata pony. She'll love it." "That's a stick with a horse head." "Which is more than most ponies give you." Bill sighed. "What about the parents?" "Not if we're gone before the candles are lit. In and out. No ball pits. No weddings." "We lost a groom at the wedding." "He ran off. Not our fault." Molly folded the map into a tight square and slid it into her jacket. Across the diner, a waitress dropped a tray of ketchup bottles. Bill flinched. Molly didn't. She was already thinking about the service corridor, the hollow cake, the six minutes that would make them rich in birthday candles and gift cards. Outside, a school bus hissed past, packed with shrieking kids in paper crowns. Bill watched them and felt his soul leave his body. "We could still rob a bank," Bill said. "Banks don't have piñatas." "Banks have guards. Parties have frosting. Frosting is easier to climb." She checked the clock. One hour until the unicorn cake arrived and the terrible plan began. Bill ordered a second coffee and prayed for a meteor. "Fine," Bill said. "But if a kid cries, I'm leaving." "Kids always cry." "Then I'm leaving immediately." Molly grabbed his sleeve. "No, you're not. I already rented the pony head." "I hate you." "You love me."
created_at
"2026-09-10T10:23:01.383Z"
9/10/2026, 10:23:01 AM
model
"deepseek/deepseek-v4.1-flash-20260910"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
161
moderation_latency
(null)
generation_time
176658
tokens_prompt
75
tokens_completion
8878
native_tokens_prompt
93
native_tokens_completion
9681
native_tokens_completion_images
(null)
native_tokens_reasoning
9002
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
0.0116451
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.20.0; linux; x64))"
http_referer
(null)
request_id
"req-1789035781-D27aALzq7qKCxl7XCn8M"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1789035781-gM8uybCtgQknpjB9yWuj"
upstream_id
"chatcmpl-RlnZz7SXQHthvo4aTDWPk1X5"
provider_responses
0
endpoint_id
"519706ce-5bef-4de6-a755-71cc3c3ac632"
id
"chatcmpl-RlnZz7SXQHthvo4aTDWPk1X5"
is_byok
false
latency
73
model_permaslug
"deepseek/deepseek-v4.1-flash-20260910"
provider_name
"DeepInfra"
status
200
total_cost
0.0116451
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
99.98%
Matches word count
n/a
neededClean
false
words
502
99.95%
Dialogue to Total Word Ratio
Ratio: 50.10%, Deviation: 0.10%
neededClean
false
wordsTotal
503
wordsDialogue
252
99.9673%