NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-16-01KHM5TZV6QC071X7R1J0E02K0
Run Details
Model:
DeepSeek V3.1
Test:
Dialogue tags
Scenario:
Write 200 words with 10% dialogue
Duration:
37.9s
Cost:
$0.000224
Provider:
DeepInfra
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 10% 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 traced a finger over the crayon-smeared invitation. “The target is fortified. Perimeter is a picket fence. Guard dog is a poodle named Mr. Snuffles, easily distracted by a well-thrown tennis ball.” Bill peered through binoculars, his expression grim. “Hostiles everywhere. Sugar-fueled, highly unpredictable. Our window is during the piñata chaos.” He shuddered. “I don’t like it. The risk is immense. One wrong move and we’re covered in sticky toddler fingerprints.” “The loot is prime,” Molly countered, tapping the schematic of the backyard. “Unwrapped, easily fenced. High concentration of plastic and batteries. We go in as clowns. It’s the perfect disguise.” “I hate clowns.” “Everyone does. It creates a perimeter of natural avoidance. We’ll be invisible.” He sighed, defeated. “Fine. But if one of those tiny terrorists cries, I’m aborting the mission.” “Deal. Just remember the plan: you create a diversion with the squeaky shoes. I’ll load the swag bag. We exit via the inflatable slide.” “This is the weirdest job we’ve ever pulled.” “And the most profitable. Now, go practice your balloon animals.”
created_at
"2026-02-16T21:32:02.927488+00:00"
2/16/2026, 9:32:02 PM
model
"deepseek/deepseek-chat-v3.1"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
982
moderation_latency
(null)
generation_time
37825
tokens_prompt
95
tokens_completion
316
native_tokens_prompt
67
native_tokens_completion
272
native_tokens_completion_images
(null)
native_tokens_reasoning
0
native_tokens_cached
64
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
0.0002238300000128
router
(null)
provider_responses
0
id
"cmpl-83e7bb93a1fc40bd833233183a1b418e"
status
200
is_byok
false
latency
50
endpoint_id
"012f7d1f-23d4-4d03-9aa3-432d8ad82757"
provider_name
"DeepInfra"
model_permaslug
"deepseek/deepseek-chat-v3.1"
api_type
"completions"
id
"gen-1771277483-FG1TbT6gEyUZkBshd3w3"
upstream_id
"cmpl-83e7bb93a1fc40bd833233183a1b418e"
total_cost
0.0002238300000128
cache_discount
0.0000051199999872
upstream_inference_cost
0
provider_name
"DeepInfra"
Evaluation details
Result
Evaluator
Details
Meta Data
6.0907%
Matches word count
n/a
neededClean
false
words
177
0.0000%
Dialogue to Total Word Ratio
Ratio: 83.62%, Deviation: 73.62%
neededClean
false
wordsTotal
177
wordsDialogue
148
3.0453%