NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-04-25-01KQ2GCCEG4245W30PNPC8KJ1E
Run Details
Model:
DeepSeek V4 Flash
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
4.9s
Cost:
$0.000091
Provider:
DeepSeek
Evaluation summary
Score
Evaluator
Details Link
60.28%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
30.1376%
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 squinted through the binoculars. “The unicorn bounce house is the key. Security’s too busy dodging sticky fingers.” Bill adjusted his fake mustache. “You want me to waltz in wearing this? I look like a depressed librarian.” “You look like a distraction,” she said. “I’ll slip the loot into the piñata bag.” “The loot? It’s a pile of Pokémon cards and a singing plushie llama. We’ll be legendarily broke.” “Don’t underestimate the secondary market for rare holographics. And that llama has a following.” He sighed. “So my job is to get tackled by a swarm of eight-year-olds?” “Precisely. Cause a scene at the cake table. Cry about the frosting. They’ll swarm you with napkins.” “I’m not crying for a unicorn cake.” “You’ll cry if we walk away empty-handed. Look, the birthday boy just dropped his own present. He’s sobbing. That’s our opening.” “That’s traumatic.” “It’s capitalism, Bill. Now go. And for heaven’s sake, straighten your mustache. You look like a caterpillar ate a sad man.” He trudged toward the party, muttering, “I’m stealing a llama for a nursery rhyme. My mother would be so proud.”
created_at
"2026-04-25T14:23:30.773Z"
4/25/2026, 2:23:30 PM
model
"deepseek/deepseek-v4-flash-20260423"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
538
moderation_latency
(null)
generation_time
4698
tokens_prompt
77
tokens_completion
328
native_tokens_prompt
67
native_tokens_completion
290
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
0.00009058
router
(null)
provider_responses
0
endpoint_id
"722e0746-2034-4d6e-a08f-e6271849ee9e"
id
"3d0c891a-d5dc-462c-a328-23662109f598"
is_byok
false
latency
280
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-1777127010-eE9NG4n9cJ3XnAmijDkV"
session_id
(null)
api_type
"completions"
id
"gen-1777127010-YHBN6RPCgZOeXLIL6cxO"
upstream_id
"3d0c891a-d5dc-462c-a328-23662109f598"
total_cost
0.00009058
cache_discount
(null)
upstream_inference_cost
0
provider_name
"DeepSeek"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
60.28%
Matches word count
n/a
neededClean
false
words
185
0.00%
Dialogue to Total Word Ratio
Ratio: 89.36%, Deviation: 39.36%
neededClean
false
wordsTotal
188
wordsDialogue
168
30.1376%