NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-10-01M25DSX13CTWXN3K7GNKF2F21
Run Details
Model:
DeepSeek V4.1 Flash (Reasoning, High)
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
2m 5s
Cost:
$0.004854
Provider:
DeepInfra
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
100.00%
Dialogue to Total Word Ratio
(details)
99.9984%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 200 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 party blueprint across the getaway tricycle. Bill sharpened a crayon and checked the guest list, which was mostly fingerprints and glitter. "We hit the party at cake o'clock." "While kids chant for sugar?" "Exactly. We slip past the clown." "The clown scares me." "Everyone fears clowns. That's leverage." "Presents are stacked by the bouncy castle." "I'll distract with balloon animals." "You can't make balloon animals." "I can make balloon snakes. Close enough." "Molly, we need precision." "Precision is for museums. This is a toddler heist." "Fine. I grab gifts. You trip the magician." "Why trip him?" "Because he's already suspicious." "He's seven." "He knows too much." "Deal. What about party favors?" "We leave those. Professional courtesy." "What about the cake?" "We eat it." Bill nodded solemnly. Molly tucked the blueprint into a diaper bag, adjusted her fake mustache, and practiced a convincing toddler waddle. Somewhere inside, forty children screamed with joy, or possibly terror. It was hard to tell. They crept toward the bouncy castle, ready to commit the pettiest crime in birthday history. The clown noticed them, sighed, and gave a weary thumbs-up, as if he had seen this exact crime three times that weekend.
created_at
"2026-09-10T10:28:51.369Z"
9/10/2026, 10:28:51 AM
model
"deepseek/deepseek-v4.1-flash-20260910"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
4427
moderation_latency
(null)
generation_time
124594
tokens_prompt
75
tokens_completion
3814
native_tokens_prompt
93
native_tokens_completion
4022
native_tokens_completion_images
(null)
native_tokens_reasoning
3737
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
0.0048543
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.20.0; linux; x64))"
http_referer
(null)
request_id
"req-1789036131-VYBJ9TFRtWtjqO5UFVRZ"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1789036131-HBAFja6KcpmmVD0aFsQW"
upstream_id
"chatcmpl-RYiSFeDR8NMfYvFqmtWcWa5u"
provider_responses
0
endpoint_id
"519706ce-5bef-4de6-a755-71cc3c3ac632"
id
"chatcmpl-RYiSFeDR8NMfYvFqmtWcWa5u"
is_byok
false
latency
76
model_permaslug
"deepseek/deepseek-v4.1-flash-20260910"
provider_name
"DeepInfra"
status
200
total_cost
0.0048543
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
100.00%
Matches word count
n/a
neededClean
false
words
200
100.00%
Dialogue to Total Word Ratio
Ratio: 50.75%, Deviation: 0.75%
neededClean
false
wordsTotal
201
wordsDialogue
102
99.9984%