NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-03-01M1KBC6W8J7PEW43BCES0EM0M
Run Details
Model:
DeepSeek V4 Flash 0731 (Reasoning, High)
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
31.1s
Cost:
$0.000132
Provider:
OpenInference
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
0.0001%
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 peered through the binoculars, her breath fogging the lenses. "There's the target. Pink bouncy castle, a piƱata shaped like a unicorn, and a table groaning under the weight of gift wrap." Bill shifted in the driver's seat, crunching a bag of chips. "Looks like a goldmine. But how do we get past the parents? They're armed with cameras and judgment." "Easy," Molly said, lowering the glasses. "We don't go in as thieves. We go in as the entertainment. I've got a clown costume in the back. Honk the nose, do a few balloon animals." Bill snorted. "You? A clown? You once scared a kid at a mall just by frowning. That's not funny, that's a trauma trigger." "Fine," she countered. "Then you go as the magician. You can pull a rabbit out of a hat while I lift the presents." "Magician? I can't even pull a quarter from behind my own ear without getting tangled. What if they ask for a trick?" "Then you do the disappearing act," Molly said, grinning. "You disappear with the loot. I'll cause a diversion by tripping over the bouncy castle." Bill sighed, tossing the empty chip bag aside. "Alright. But I'm keeping the cone hat. It's stylish." Molly smiled. "Deal. Now, go put on the wig. And remember, smile like you actually like kids." Bill grumbled, opening the van door. "I'll smile when I'm counting the cash."
created_at
"2026-09-03T10:00:05.804Z"
9/3/2026, 10:00:05 AM
model
"deepseek/deepseek-v4-flash-20260731"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
503
moderation_latency
(null)
generation_time
31019
tokens_prompt
75
tokens_completion
810
native_tokens_prompt
67
native_tokens_completion
807
native_tokens_completion_images
(null)
native_tokens_reasoning
473
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
0.00013247
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788429605-N8ikqzW7CAdkfbKd6xuO"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788429605-yaXnpdj9cuqUjit68hpf"
upstream_id
"chatcmpl-dde06e9c-16a5-41b5-a3ef-eb13c7a9a565"
provider_responses
0
endpoint_id
"f954a48e-1c90-433b-9348-4720f8030331"
id
"chatcmpl-dde06e9c-16a5-41b5-a3ef-eb13c7a9a565"
is_byok
false
latency
503
model_permaslug
"deepseek/deepseek-v4-flash-20260731"
provider_name
"OpenInference"
status
200
total_cost
0.00013247
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
0.00%
Matches word count
n/a
neededClean
false
words
234
0.00%
Dialogue to Total Word Ratio
Ratio: 79.06%, Deviation: 29.06%
neededClean
false
wordsTotal
234
wordsDialogue
185
0.0001%