NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-04-25-01KQ2E05KNPG0407DA519V4KB9
Run Details
Model:
DeepSeek V4 Flash (Reasoning)
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
5.9s
Cost:
$0.000122
Provider:
DeepSeek
Evaluation summary
Score
Evaluator
Details Link
43.38%
Matches word count
(details)
98.27%
Dialogue to Total Word Ratio
(details)
70.8225%
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
“So you’re telling me we’re robbing a birthday party for six-year-olds?” Bill adjusted his balaclava, which was frankly too small. “I thought we agreed on the diamond exchange.” “The diamond exchange has lasers, Bill. Toddlers have juice boxes and a bouncy castle.” Molly scanned the backyard with her binoculars. “Look at that pile. Unwrapped. Unwatched. It’s a goldmine of plastic, but gold is gold.” “It’s a goldmine of Easy-Bake Ovens. We don’t bake.” “We fence. Karen’s cousin runs a Facebook group, ‘Pre-Loved Toys, No Lowballers.’ We’ll make bank.” “I’m not crawling through a ball pit for a Furby that probably has fleas.” “It’s not fleas, it’s ‘batteries not included.’” She handed him a pink party hat. “Wear this. Blend in.” “I look like a perverted unicorn.” “You look like Uncle Steve. Now, the parents are inside doing the cake. We have exactly seven minutes. You take the north end of the present mountain, I’ll sweep the south. Grab anything with a tag from Granny.” “And if a kid sees us?” “Smile and say ‘Happy Birthday.’ They’re three feet tall. They can’t identify you in a lineup.” Bill sighed. “I’m going to get glitter on my good crowbar.” “That’s the spirit.” Molly crept over the fence. “Remember, silent and swift.” “I’ve stepped on a Lego. I’m already compromised.”
created_at
"2026-04-25T13:41:53.401Z"
4/25/2026, 1:41:53 PM
model
"deepseek/deepseek-v4-flash-20260423"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
523
moderation_latency
(null)
generation_time
5762
tokens_prompt
77
tokens_completion
443
native_tokens_prompt
67
native_tokens_completion
404
native_tokens_completion_images
(null)
native_tokens_reasoning
52
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
0.0001225
router
(null)
provider_responses
0
endpoint_id
"722e0746-2034-4d6e-a08f-e6271849ee9e"
id
"ab2237dd-e8d0-4a69-9eb6-6ebcb94b4f2c"
is_byok
false
latency
401
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-1777124513-2zPlOK22MLizL29jDJHR"
session_id
(null)
api_type
"completions"
id
"gen-1777124513-CjCt22SEvhVPmEfUlNdb"
upstream_id
"ab2237dd-e8d0-4a69-9eb6-6ebcb94b4f2c"
total_cost
0.0001225
cache_discount
(null)
upstream_inference_cost
0
provider_name
"DeepSeek"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
43.38%
Matches word count
n/a
neededClean
false
words
217
98.27%
Dialogue to Total Word Ratio
Ratio: 86.36%, Deviation: 3.64%
neededClean
false
wordsTotal
220
wordsDialogue
190
70.8225%