NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-04-24-01KQ0NQ2RGHKFNA8NMHMNZ6DPH
Run Details
Model:
GPT-5.5
Test:
Dialogue tags
Scenario:
Write 200 words with 10% dialogue
Duration:
7.5s
Cost:
$0.008160
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
99.38%
Matches word count
(details)
91.94%
Dialogue to Total Word Ratio
(details)
95.6570%
Overall
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
Molly crouched behind the hydrangeas, wearing a caterpillar costume that smelled faintly of attic and regret. Across the street, the Henderson house glowed with balloons, streamers, and the unmistakable shriek of sugared children. Bill unfolded a blueprint he had drawn on the back of a pizza menu, then held it upside down with professional confidence. Inside, a magician produced scarves from his sleeve while parents applauded with the dead eyes of the defeated. Near the fireplace, a mountain of presents shimmered in dinosaur paper and smug ribbons. “Cake first, loot second,” Molly whispered. Bill nodded gravely. “No, presents first. Cake distracts.” A golden retriever in a party hat spotted them and began wagging like a security alarm made of joy. Molly tossed it a sausage roll. The dog accepted the bribe and joined their criminal enterprise. Their plan was simple: enter as entertainers, perform one balloon animal, create confusion, evacuate gifts in a laundry cart labeled Extremely Boring Towels. Unfortunately, Bill’s balloon animal was a sword, his only trick, and the birthday boy immediately challenged him to a duel. Molly sighed, adjusted her antennae, and rolled the cart toward destiny, frosting, and probable community service.
created_at
"2026-04-24T21:18:15.319Z"
4/24/2026, 9:18:15 PM
model
"openai/gpt-5.5-20260423"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
744
moderation_latency
(null)
generation_time
7402
tokens_prompt
77
tokens_completion
328
native_tokens_prompt
66
native_tokens_completion
261
native_tokens_completion_images
0
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
"completed"
usage
0.00816
router
(null)
provider_responses
0
endpoint_id
"58e5b336-423e-430b-a2ab-8bc353f0c51b"
id
"resp_07e1274f95df6b330169ebde176ac88194af5d85e6fc45c544"
is_byok
false
latency
413
model_permaslug
"openai/gpt-5.5-20260423"
provider_name
"OpenAI"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.14.1; linux; x64))"
http_referer
(null)
request_id
"req-1777065495-HTjd7l8YjtqsUWpVrEHl"
session_id
(null)
api_type
"completions"
id
"gen-1777065495-ys5VfEEQ3DZCMUdfdONv"
upstream_id
"resp_07e1274f95df6b330169ebde176ac88194af5d85e6fc45c544"
total_cost
0.00816
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
99.38%
Matches word count
n/a
neededClean
false
words
195
91.94%
Dialogue to Total Word Ratio
Ratio: 4.62%, Deviation: 5.38%
neededClean
false
wordsTotal
195
wordsDialogue
9
95.6570%