NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-03-01M1K9KYT34DARZS74NMRSZMGG
Run Details
Model:
DeepSeek V4 Flash 0731 (Reasoning, High)
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
28.7s
Cost:
$0.000124
Provider:
OpenInference
Evaluation summary
Score
Evaluator
Details Link
20.19%
Matches word count
(details)
99.64%
Dialogue to Total Word Ratio
(details)
59.9166%
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
Molly adjusted her ski mask. "Bill, the target is the birthday party at the FunZone annex. The haul: a mountain of wrapped Lego, a Furby, and three gift cards to Build-a-Bear." Bill squinted at the blueprint, which was a crayon drawing on a napkin. "And the security system? A five-year-old with a juice box?" "Worse," Molly said. "Her name is Charlotte. The birthday girl. She has a whistle. And a mother who does not mess around." "A whistle? That's our kryptonite," Bill sighed. "How do we neutralize the whistle?" "Use the distraction," Molly said. "I'll trip over the balloon arch. You sneak in through the bounce house's emergency exit." "Bounce house. Very high-profile. What if I get stuck?" "You'll be fine. Slim build. The real danger is the mascot. Chuck E. himself is an ex-con named Vinny." "Vinny? The guy who did time for stealing a piñata?" "The same. He's onto us. He knows we're coming." "How?" "Because I told him. He wants a cut. Forty percent." "Forty percent! That's extortion!" "It's business, Bill. He also wants the Furby. Has a thing for its creepy eyes." Bill shook his head. "Fine. But if Charlotte drops a beat, I'm out." "Deal. Let's move. Whatever you do, don't step on the cake." "Why?" "It's the decoy. The presents are in the piñata."
created_at
"2026-09-03T09:29:22.62Z"
9/3/2026, 9:29:22 AM
model
"deepseek/deepseek-v4-flash-20260731"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
570
moderation_latency
(null)
generation_time
27838
tokens_prompt
75
tokens_completion
748
native_tokens_prompt
67
native_tokens_completion
752
native_tokens_completion_images
(null)
native_tokens_reasoning
417
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
0.00012367
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788427763-dLEreSOehoE4hflmd29p"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788427763-5yN54i2sXcwxYxsgwy7F"
upstream_id
"chatcmpl-50b6e171-2c46-4895-a0d6-9b2281619e48"
provider_responses
0
endpoint_id
"f954a48e-1c90-433b-9348-4720f8030331"
id
"chatcmpl-50b6e171-2c46-4895-a0d6-9b2281619e48"
is_byok
false
latency
570
model_permaslug
"deepseek/deepseek-v4-flash-20260731"
provider_name
"OpenInference"
status
200
total_cost
0.00012367
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenInference"
response_cache_source_id
(null)
data_region
"global"
workspace_id
"97e315e5-d303-487d-83c1-83180e8a13d4"
Evaluation details
Result
Evaluator
Details
Meta Data
20.19%
Matches word count
n/a
neededClean
false
words
220
99.64%
Dialogue to Total Word Ratio
Ratio: 87.56%, Deviation: 2.44%
neededClean
false
wordsTotal
225
wordsDialogue
197
59.9166%