NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-16-01KHKAG4BDQEQHNMQ1PV32VXG3
Run Details
Model:
GPT-5.2
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
41.8s
Cost:
$0.035521
Provider:
OpenAI
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
“Molly, why are we robbing a children’s party?” Bill asked. “Because the presents are pre-wrapped, and nobody expects a getaway van shaped like an ice-cream truck.” “I don’t have an ice-cream truck.” “You do now. I stole it from a clown.” “At least the clown won’t file a report.” “Remember the aquarium job?” “The one where you tried to bribe a shark?” “So, plan?” “We arrive during ‘Pin the Tail,’ masks already on. Everyone thinks we’re entertainers.” “My mask is a ski mask.” “Call it a modern mime.” “And the loot?” “Gift table. We sweep it like hungry librarians. You bag, I distract.” “How do you distract six-year-olds?” “I announce a surprise game: ‘Guess Which Adult Isn’t a Parent.’” “They’ll guess us.” “Exactly. While they point and scream, you load presents into the truck.” “What about the parents?” “Parents will be filming, not thinking. I’ll hand out cupcakes; sugar is the great equalizer.” “Exit route?” “Through the bounce house. Nobody can chase you with dignity.” “And if we’re caught?” “We say, ‘It’s a magic trick,’ then vanish.” “How do we vanish?” “Smoke bombs and a piñata.” Molly grinned. “We hit at three.” Bill sighed. “Fine, but I’m keeping the piñata.”
created_at
"2026-02-16T13:34:19.377941+00:00"
2/16/2026, 1:34:19 PM
model
"openai/gpt-5.2-20251211"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
5588
moderation_latency
(null)
generation_time
41587
tokens_prompt
77
tokens_completion
975
native_tokens_prompt
66
native_tokens_completion
2529
native_tokens_completion_images
0
native_tokens_reasoning
2232
native_tokens_cached
0
num_media_prompt
(null)
num_input_audio_prompt
(null)
num_media_completion
0
num_search_results
(null)
origin
"https://ncbench.com/"
is_byok
false
finish_reason
"stop"
native_finish_reason
"completed"
usage
0.0355215
router
(null)
provider_responses
0
id
"resp_0e514a9bc0fcd87d0169931cb093a881958e914871f3640935"
status
200
is_byok
false
latency
151
endpoint_id
"f00142c2-6a93-49ce-9e36-5593b904ce3b"
provider_name
"OpenAI"
model_permaslug
"openai/gpt-5.2-20251211"
api_type
"completions"
id
"gen-1771248816-1Mse9P7JBLTM1pB1AsSk"
upstream_id
"resp_0e514a9bc0fcd87d0169931cb093a881958e914871f3640935"
total_cost
0.0355215
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
Evaluation details
Result
Evaluator
Details
Meta Data
99.7443%
Matches word count
n/a
neededClean
false
words
204
78.0147%
Dialogue to Total Word Ratio
Ratio: 97.06%, Deviation: 7.06%
neededClean
false
wordsTotal
204
wordsDialogue
198
88.8795%