NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-04-25-01KQ2F7B4PBEAWVJH2QWTXGGNN
Run Details
Model:
DeepSeek V4 Flash
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
5.1s
Cost:
$0.000101
Provider:
DeepSeek
Evaluation summary
Score
Evaluator
Details Link
6.09%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
3.0453%
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 blueprints across the sticky diner table. "Okay, Bill. The kid’s party runs from 2 to 5. We go in as 'Uncle Frank' and 'Fun Aunt Carol.'" Bill poked at a jelly doughnut. "Fun Aunt Carol? You hate children." "I hate losing," she said. "The gift table is by the bouncy castle. You create a diversion." "What kind of diversion?" "Pop the bouncy castle. Slowly. The kids will be furious. Parents will panic." Bill frowned. "I don't pop things. Last time I popped a piñata, a kid cried for an hour. I gave him my watch." "Fine. You drop the cake. I'll take the heat." "Drop the cake? That's war, Molly. Bakeries have feelings." "The baker won't be there, you softie. Just the grandma who spent forty dollars on a SpongeBob sheet cake." Bill sighed. "Why can't we steal something normal? Like purses." "Because we're artists, Bill. Artists don't steal wallets. We steal a unicorn-themed loot bag filled with candy and a third-party tablet. Swing by the driveway. I'll fake a sprained ankle." "You do a great sprain," Bill admitted. "I practice in the mirror. Now remember—if the birthday boy asks, we're celebrating your vasectomy. He won't know what it means, but the adults will be too awkward to question it." Bill nodded slowly. "This is why you're the mastermind."
created_at
"2026-04-25T14:03:17.018Z"
4/25/2026, 2:03:17 PM
model
"deepseek/deepseek-v4-flash-20260423"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
522
moderation_latency
(null)
generation_time
5006
tokens_prompt
77
tokens_completion
345
native_tokens_prompt
67
native_tokens_completion
327
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)
num_fetches
(null)
web_search_engine
(null)
origin
"https://ncbench.com/"
is_byok
false
finish_reason
"stop"
native_finish_reason
"stop"
usage
0.00010094
router
(null)
provider_responses
0
endpoint_id
"722e0746-2034-4d6e-a08f-e6271849ee9e"
id
"9a3d34ed-a3dd-4099-9119-e05167237d77"
is_byok
false
latency
278
model_permaslug
"deepseek/deepseek-v4-flash-20260423"
provider_name
"DeepSeek"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.14.1; linux; x64))"
http_referer
(null)
request_id
"req-1777125797-5nUhnHgh86R0qnv1Ulgc"
session_id
(null)
api_type
"completions"
id
"gen-1777125797-lOuJzybU9PpikuqBrGHj"
upstream_id
"9a3d34ed-a3dd-4099-9119-e05167237d77"
total_cost
0.00010094
cache_discount
(null)
upstream_inference_cost
0
provider_name
"DeepSeek"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
6.09%
Matches word count
n/a
neededClean
false
words
223
0.00%
Dialogue to Total Word Ratio
Ratio: 88.44%, Deviation: 38.44%
neededClean
false
wordsTotal
225
wordsDialogue
199
3.0453%