NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-10-20-01M5C6FMYH3B5TYATJDNJX08D6
Run Details
Model:
DeepSeek V4.1 Flash 1017 (Reasoning, High)
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
23.7s
Cost:
$0.002065
Provider:
InferenceNet
Evaluation summary
Score
Evaluator
Details Link
0.08%
Matches word count
(details)
99.60%
Dialogue to Total Word Ratio
(details)
49.8448%
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 crouched behind the hedge, binoculars pressed to her eyes, while Bill wrestled a clown costume out of a duffel bag. A bouncy castle wheezed in the backyard, and a pack of six-year-olds swarmed a piƱata shaped like a unicorn. "Remind me again why we're robbing a birthday party," Bill whispered. "Because the Richmond job fell through and this party has a gift table with more boxes than a shipping depot," Molly said. "Also the birthday girl's uncle is a jeweler and he brought the good stuff." Bill pulled the clown wig over his head. "I look like a tax auditor who got mugged by a cotton candy machine." "You look perfect," Molly said. "The plan is simple. You distract the kids with balloon animals. I slip behind the table and load the presents into the stroller." "Simple? There's a bouncy castle and a petting zoo." "The goat is not security, Bill." "The goat ate my glove at the last party." Molly checked her watch. "We go in three minutes. Did you learn the balloon dog?" Bill inflated a long red balloon. It burst with a sound like a gunshot. In the yard a dozen children froze then screamed with delight and charged toward the hedge. "Great," Molly hissed. "Now you've started a stampede." "I panicked," Bill said. "That wasn't a dog. That was a distress signal." A small boy in a pirate hat pointed at Bill. "Are you a clown?" "No," Bill said. "I'm a highly trained balloon artist with a criminal record." Molly pinched the bridge of her nose. "He means yes. Yes he is a clown." The boy squinted. "Why does the clown smell like motor oil?" "Professional hazard," Bill said. "Do you want a sword?" He twisted a green balloon into something resembling a bent banana. The boy took it and immediately hit another child. "Excellent," Molly said. "Now we have a riot." She pushed the stroller toward the gift table but a woman in a party hat stepped in front of her. "Are you with the entertainment?" Molly smiled. "I'm the clown's manager." "He's terrible." "He's avant-garde." The woman frowned. "Why is he juggling the cake?" Bill had indeed lifted the birthday cake and was attempting a three-item cascade with a juice box and a plastic dinosaur. "That's part of the act," Molly said. "It's called chaos theory." The cake tilted. Bill caught it with his face. The children cheered. Molly grabbed three wrapped presents and shoved them under the stroller. "We need to go," she whispered. "Not yet," Bill said wiping frosting from his eyes. "The goat wants a balloon hat." "Bill the police aren't coming. The parents are." A group of fathers advanced with napkins and wet wipes. Molly yanked Bill by the collar. "Run." "Can I keep the wig?" "No." "Can I keep the dinosaur?" "No." They bolted through the gate the stroller rattling a small pirate chasing them with a balloon sword. "This," Bill panted, "is why I prefer museums." "At least kids don't call the cops," Molly said. "They call their mothers." Behind them a chorus of tiny voices rose. "Mommy! The clown stole my present!" "Faster," Bill said.
created_at
"2026-10-20T08:22:41.878Z"
10/20/2026, 8:22:41 AM
model
"deepseek/deepseek-v4.1-flash-20261017"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
116
moderation_latency
(null)
generation_time
23659
tokens_prompt
75
tokens_completion
3784
native_tokens_prompt
67
native_tokens_completion
4120
native_tokens_completion_images
(null)
native_tokens_reasoning
3392
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
0.00206536
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.21.0; linux; x64))"
http_referer
(null)
request_id
"req-1792484561-xiWrxSBs84xYoUgLyt5O"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1792484561-6nDyHaUwvFqJ0VWnSa39"
upstream_id
"gv2_0Oiliju4KzIBjeZt4ymkF"
provider_responses
0
endpoint_id
"4231aea9-73af-4a2f-bd14-660b1d8b13b2"
id
"gv2_0Oiliju4KzIBjeZt4ymkF"
is_byok
false
latency
116
model_permaslug
"deepseek/deepseek-v4.1-flash-20261017"
provider_name
"InferenceNet"
status
200
total_cost
0.00206536
cache_discount
(null)
upstream_inference_cost
0
provider_name
"InferenceNet"
response_cache_source_id
(null)
data_region
"global"
workspace_id
"97e315e5-d303-487d-83c1-83180e8a13d4"
Evaluation details
Result
Evaluator
Details
Meta Data
0.08%
Matches word count
n/a
neededClean
false
words
529
99.60%
Dialogue to Total Word Ratio
Ratio: 49.72%, Deviation: 0.28%
neededClean
false
wordsTotal
533
wordsDialogue
265
49.8448%