NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-03-01M1KBS6FH3Z7VKWJCZ30TPNM4
Run Details
Model:
DeepSeek V4 Pro 0813 (Reasoning, High)
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
10.1s
Cost:
$0.001522
Provider:
Together
Evaluation summary
Score
Evaluator
Details Link
99.38%
Matches word count
(details)
36.79%
Dialogue to Total Word Ratio
(details)
68.0824%
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
"Bill, I've cased the bouncy castle. The presents are stacked behind the piñata, guarded by a woman in a cupcake apron." "Cupcake apron? Armed with frosting?" "Worse. She's got a whistle. One blow and forty sugar-crazed toddlers swarm us." "So we go in as clowns." "Too obvious. I'm thinking magician and assistant. I pull a rabbit out of a hat, you load the gift bags into the false-bottomed wagon." "Where do we get a rabbit?" "Stuffed. The kids won't know the difference. They're four." "Fine. But the birthday boy's mother knows my face from that bounce house incident in July." "Wear the rainbow wig. And stop making eye contact with the pony." "There's a pony?" "Focus, Bill. At exactly 2:15, the clown they hired does balloon animals. That's our window. I'll distract with a card trick, you grab the big box with the red ribbon." "Why the red ribbon?" "Because it's heavy. Heavy means expensive. Probably a tablet or a tiny dirt bike." "Or a bag of encyclopedias." "Nobody gives encyclopedias anymore. It's a drone. I can feel it." "Fine. But if the piñata bursts, we abort. I'm not getting beaten with a foam sword again." "Deal. Now practice your 'ta-da' face. You look constipated."
created_at
"2026-09-03T10:07:11.363Z"
9/3/2026, 10:07:11 AM
model
"deepseek/deepseek-v4-pro-20260813"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
910
moderation_latency
(null)
generation_time
10038
tokens_prompt
75
tokens_completion
369
native_tokens_prompt
67
native_tokens_completion
362
native_tokens_completion_images
(null)
native_tokens_reasoning
51
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
0.00152196
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788430031-ukrHdyOMgY4wctuesQtt"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788430031-LYdB77PQzWNvqr8XTcol"
upstream_id
"2b5b2a5f-aws_ue2"
provider_responses
0
endpoint_id
"218b0cec-3278-4b88-8ee0-3f346eccd35f"
id
"2b5b2a5f-aws_ue2"
is_byok
false
latency
910
model_permaslug
"deepseek/deepseek-v4-pro-20260813"
provider_name
"Together"
status
200
total_cost
0.00152196
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Together"
response_cache_source_id
(null)
data_region
"global"
workspace_id
"97e315e5-d303-487d-83c1-83180e8a13d4"
Evaluation details
Result
Evaluator
Details
Meta Data
99.38%
Matches word count
n/a
neededClean
false
words
205
36.79%
Dialogue to Total Word Ratio
Ratio: 100.00%, Deviation: 10.00%
neededClean
false
wordsTotal
208
wordsDialogue
208
68.0824%