NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-03-01M1K9TWRPBPKDXPV6Y4TBD9DC
Run Details
Model:
DeepSeek V4 Flash 0731 (Reasoning, High)
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
15.3s
Cost:
$0.000079
Provider:
Relace
Evaluation summary
Score
Evaluator
Details Link
99.92%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
49.9595%
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 flattened the party invitation against the van’s dashboard. “We hit at three, right when the clown starts the balloon animals. That’s when the parents are distracted.” Bill squinted at the backyard layout. “And the presents go under the big oak tree. You said there’s a Labrador.” “Jellybean. He’ll bark at the bounce house, not us. We go over the fence, through the hedges, and we’re at the gift table in forty seconds.” “Forty seconds,” Bill repeated. “What about the birthday girl?” “She’ll be crying. The clown’s never funny. Trust me.” Bill chewed his lip. “Stealing from a five-year-old feels like a new low, Moll.” “Her dad’s a plastic surgeon. These are the good presents. Think of it as adult adoption.” “Adoption,” Bill muttered. “Right.” Molly nudged him. “You’re the one who wanted a real heist. This is the realest heist of your life. No vaults, no lasers, just a bounce house and a weeping clown.” Bill sighed, then grinned. “I’ll bring the duffel bag.” “And the decoy confetti cannons?” “Of course.” Molly folded the invitation. “Then we’re set. Let’s go make some kid’s memories slightly less grand.” “That’s the spirit,” Bill said. “Let’s rob a toddler.”
created_at
"2026-09-03T09:33:09.796Z"
9/3/2026, 9:33:09 AM
model
"deepseek/deepseek-v4-flash-20260731"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
596
moderation_latency
(null)
generation_time
14293
tokens_prompt
75
tokens_completion
417
native_tokens_prompt
146
native_tokens_completion
388
native_tokens_completion_images
(null)
native_tokens_reasoning
76
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
0.00007933
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.19.0; linux; x64))"
http_referer
(null)
request_id
"req-1788427989-O2T7ROhTXc1q5syDqsjD"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1788427989-a3gVd8b7dCxYriwxWdcE"
upstream_id
"1a646f88cfc34e4b84b0773c6b23dba1"
provider_responses
0
endpoint_id
"d880a270-04ff-479c-b312-6246c0f985b0"
is_byok
false
latency
364
model_permaslug
"deepseek/deepseek-v4-flash-20260731"
provider_name
"StreamLake"
status
429
1
endpoint_id
"cb029acf-a84f-4280-8103-909d269e1e5f"
is_byok
false
latency
577
model_permaslug
"deepseek/deepseek-v4-flash-20260731"
provider_name
"DigitalOcean"
status
429
2
endpoint_id
"57c1bfab-049c-4d6a-ab34-ac1007a6043b"
id
"1a646f88cfc34e4b84b0773c6b23dba1"
is_byok
false
latency
595
model_permaslug
"deepseek/deepseek-v4-flash-20260731"
provider_name
"Relace"
status
200
total_cost
0.00007933
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Relace"
response_cache_source_id
(null)
data_region
"global"
workspace_id
"97e315e5-d303-487d-83c1-83180e8a13d4"
Evaluation details
Result
Evaluator
Details
Meta Data
99.92%
Matches word count
n/a
neededClean
false
words
197
0.00%
Dialogue to Total Word Ratio
Ratio: 81.91%, Deviation: 31.91%
neededClean
false
wordsTotal
199
wordsDialogue
163
49.9595%