NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-04-24-01KQ0N893QPR8HQ5TXM0AFN010
Run Details
Model:
GPT-5.5 (Reasoning)
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
53.4s
Cost:
$0.113880
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
36.79%
Dialogue to Total Word Ratio
(details)
68.3940%
Overall
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, confirm the objective.” “Bill, we infiltrate the birthday party, liberate the presents.” “Liberate sounds noble.” “Exactly. We’re practically elves with worse paperwork.” “The invitation says ‘Princess Jasper’s Unicorn Jubilee.’” “Then wear the horn.” “This is a plunger.” “Confidence makes it royal.” “What’s our cover?” “Children’s entertainers.” “Our act?” “You juggle.” “I have three oranges and a fear of applause.” “Good. Vulnerability plays well.” “And you?” “I do magic.” “You know magic?” “I can make suspicion disappear by offering stickers.” “Brilliant. What about parents?” “Distract them with the phrase ‘nut-free cake policy.’” “That starts conversations?” “That starts committees.” “What about the gift table?” “Behind the cupcake tower, guarded by a grandmother.” “Dangerous.” “Deadly. Smile and ask about her knee.” “Then?” “Swap presents for decoys.” “Molly, the decoys are potatoes.” “Educational toys. They teach disappointment.” “Some child will cry.” “Then cake.” “Bill, are you going soft?” “I’m going glittery. Vest sheds.” “Focus. We need that remote-control dinosaur.” “For profit?” “For dignity. Goose chased us.” “You still owe that goose an apology.” “I owe it nothing.” “Abort if anyone sings?” “Never. We harmonize, blend in, grab dinosaur, exit through the bouncy castle.” “That’s not an exit.” “With determination, everything is an exit.”
created_at
"2026-04-24T21:10:10.306Z"
4/24/2026, 9:10:10 PM
model
"openai/gpt-5.5-20260423"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
50217
moderation_latency
(null)
generation_time
53182
tokens_prompt
77
tokens_completion
692
native_tokens_prompt
66
native_tokens_completion
3785
native_tokens_completion_images
0
native_tokens_reasoning
3467
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
0.11388
router
(null)
provider_responses
0
endpoint_id
"58e5b336-423e-430b-a2ab-8bc353f0c51b"
id
"resp_0621d3535fcc62fc0169ebdc3264188196b5b6edc21f957a4d"
is_byok
false
latency
360
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-1777065010-HF3bjRHUtTfcEGi1rFSD"
session_id
(null)
api_type
"completions"
id
"gen-1777065010-l6veLkztlzYXAcWhbvHF"
upstream_id
"resp_0621d3535fcc62fc0169ebdc3264188196b5b6edc21f957a4d"
total_cost
0.11388
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
100.00%
Matches word count
n/a
neededClean
false
words
200
36.79%
Dialogue to Total Word Ratio
Ratio: 100.00%, Deviation: 10.00%
neededClean
false
wordsTotal
201
wordsDialogue
201
68.3940%