NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-16-01KHM63FR2QGM8C4EYWVB2CTNC
Run Details
Model:
DeepSeek V3.2
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
9.1s
Cost:
$0.000153
Provider:
Novita
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 90% 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
“Okay, the intel’s solid. The target’s a seven-year-old’s birthday, tomorrow at three,” Molly said, spreading a crayon-drawn map on the table. Bill squinted. “This is a dinosaur. A triceratops.” “It’s the backyard! The head is the slide, the legs are the picnic tables. Focus. Parents will be distracted by a magician. We go in as entertainers.” “We’re not entertainers.” “We’re ‘The Balloon Buddies.’ You twist, I’ll grab. We hit the present table by the ‘frill’—that’s the hedge.” “What’s my skill set here, Mol? I can barely tie my shoes.” “You’ll learn. YouTube. It’s all long balloons and simple dogs. The loot is wrapped, small, and plentiful. We fill the giant ‘duffel bag’ balloon bouquet bag.” Bill rubbed his temples. “The exit strategy?” “We leave the balloons, cause a diversion—you’ll ‘accidentally’ pop the giant piñata early. Chaos ensues. We walk out the side gate, cool as cucumbers.” “I hate cucumbers.” “Figure of speech! Look, these are luxury toys. That new mini-gold-drone-thing? It’s in there. It’s a sweet score.” “Fine. But if one snot-nosed kid asks for a giraffe, we’re blown.” “Deal. Now practice. That’s a poodle, Bill. That’s a… distressed worm. Keep practicing.”
created_at
"2026-02-16T21:36:11.804323+00:00"
2/16/2026, 9:36:11 PM
model
"deepseek/deepseek-v3.2-20251201"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1062
moderation_latency
(null)
generation_time
9048
tokens_prompt
77
tokens_completion
361
native_tokens_prompt
67
native_tokens_completion
337
native_tokens_completion_images
(null)
native_tokens_reasoning
0
native_tokens_cached
0
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
0.000152823
router
(null)
provider_responses
0
id
"2736d2e9a6f39794c148600e5e6778a0"
status
200
is_byok
false
latency
1062
endpoint_id
"8a5fb6d3-ef71-4958-af54-269ffa3a3c6e"
provider_name
"Novita"
model_permaslug
"deepseek/deepseek-v3.2-20251201"
api_type
"completions"
id
"gen-1771277762-CP1gthAOnryiTaOoeL6j"
upstream_id
"2736d2e9a6f39794c148600e5e6778a0"
total_cost
0.000152823
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Novita"
Evaluation details
Result
Evaluator
Details
Meta Data
100.0000%
Matches word count
n/a
neededClean
false
words
200
99.8401%
Dialogue to Total Word Ratio
Ratio: 92.00%, Deviation: 2.00%
neededClean
false
wordsTotal
200
wordsDialogue
184
99.9201%