NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-10-20-01M5C6PCCPDD381T6SR22KHWA8
Run Details
Model:
DeepSeek V4.1 Flash 1017 (Reasoning, High)
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
12.9s
Cost:
$0.000861
Provider:
InferenceNet
Evaluation summary
Score
Evaluator
Details Link
75.16%
Matches word count
(details)
68.26%
Dialogue to Total Word Ratio
(details)
71.7074%
Overall
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, Bill, the clown gets the kids to the piñata at two, so we’ve got six minutes." "That’s four minutes to snatch the presents and two to argue about who took the unicorn," Bill said. "I’m not arguing. I’m establishing custody." "Fine. You take the wrapped boxes, I’ll take the gift bags." "Why do you get gift bags?" "Because they’re lighter, and you owe me for the hamster incident." "We agreed never to mention the hamster." "Then stop making sense. The back door’s by the cake table?" "Yes, but the birthday kid’s mom guards it like a dragon with a clipboard." "So we go through the bouncy castle." "It’s inflatable, Bill." "Exactly. No one suspects a tactical bounce." "Roll call: masks, sacks, distraction?" "I brought a kazoo." "A kazoo?" "It’s terrifying in the right hands." "Your hands aren’t right. Give it here." "Careful, Molly. That’s our retirement plan." "No, that’s a plastic horn. The retirement plan is the princess castle box." "Then let’s go steal childhood joy." "Also, if a toddler asks what we’re doing, we’re magicians." "Bad magicians." "Method actors." "Finally, a noble cause," Molly said.
created_at
"2026-10-20T08:26:22.491Z"
10/20/2026, 8:26:22 AM
model
"deepseek/deepseek-v4.1-flash-20261017"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
169
moderation_latency
(null)
generation_time
12852
tokens_prompt
75
tokens_completion
1650
native_tokens_prompt
67
native_tokens_completion
1711
native_tokens_completion_images
(null)
native_tokens_reasoning
1415
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
0.00086086
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.21.0; linux; x64))"
http_referer
(null)
request_id
"req-1792484782-kWc3MCSzhWEn3O6cTTfQ"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1792484782-NM5lMjD139YfL1vtYbxX"
upstream_id
"gv2_oEoOpSdZehPKmUf7KXP0V"
provider_responses
0
endpoint_id
"4231aea9-73af-4a2f-bd14-660b1d8b13b2"
id
"gv2_oEoOpSdZehPKmUf7KXP0V"
is_byok
false
latency
169
model_permaslug
"deepseek/deepseek-v4.1-flash-20261017"
provider_name
"InferenceNet"
status
200
total_cost
0.00086086
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
75.16%
Matches word count
n/a
neededClean
false
words
187
68.26%
Dialogue to Total Word Ratio
Ratio: 97.86%, Deviation: 7.86%
neededClean
false
wordsTotal
187
wordsDialogue
183
71.7074%