NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-16-01KHKB0GQ50CM8JEJQR9CK2ZDX
Run Details
Model:
GPT-5 Nano
Test:
Dialogue tags
Scenario:
Write 200 words with 10% dialogue
Duration:
2m 2s
Cost:
$0.005573
Provider:
OpenAI
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 10% 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
Under the glow of a streetlamp, Molly and Bill studied the map of the neighborhood like two eccentric chefs plotting a bake sale disaster. A banner for a child's party fluttered in the wind, and plastic bells rattled from a nearby garage. Molly adjusted her tiny backpack, which held a flashlight, a crowbar, and a thorough misunderstanding of social norms. Bill squinted through his night-vision goggles, which were really just blue-tinted sunglasses with a receipt stuck to them. The plan crawled out of their brains like a stage whisper: big, ridiculous, and perfectly impractical. A whispered calendar counted down to the moment when the party would sigh with sugar and laughter. They spoke in half-shouted bungled sentences, testing bravado against the sound of a children's chorus rehearsing Jingle Bells. "Molly: We strike after the cake." "Bill: Only after the balloons." "Molly: Not the games, the presents." "Bill: Fine, then." Outside, a stray cat treated the scene like a stage, tail flicking to the beat. Molly checked the perimeter, muttering about doormen and detours, while Bill rehearsed his surrender face for when the party favors finally awoke. They grinned, confident only in their ridiculous brilliance. The plan shimmered, ridiculous as always.
created_at
"2026-02-16T13:44:36.318841+00:00"
2/16/2026, 1:44:36 PM
model
"openai/gpt-5-nano-2025-08-07"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
3739
moderation_latency
(null)
generation_time
121909
tokens_prompt
77
tokens_completion
1470
native_tokens_prompt
66
native_tokens_completion
13925
native_tokens_completion_images
0
native_tokens_reasoning
13632
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
0.0055733
router
(null)
provider_responses
0
id
"resp_04ea6d9f5aa846710169931ec984608193b928f057e7f3b2be"
status
200
is_byok
false
latency
107
endpoint_id
"50329d77-04e1-4979-a184-c33030289476"
provider_name
"OpenAI"
model_permaslug
"openai/gpt-5-nano-2025-08-07"
api_type
"completions"
id
"gen-1771249353-iKbonYtbNtDusml1Igs2"
upstream_id
"resp_04ea6d9f5aa846710169931ec984608193b928f057e7f3b2be"
total_cost
0.0055733
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
Evaluation details
Result
Evaluator
Details
Meta Data
99.9190%
Matches word count
n/a
neededClean
false
words
203
100.0000%
Dialogue to Total Word Ratio
Ratio: 9.85%, Deviation: 0.15%
neededClean
false
wordsTotal
203
wordsDialogue
20
99.9595%