NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-04-24-01KQ0N6S5KDEJ295Y88H0J2WF2
Run Details
Model:
GPT-5.5 (Reasoning, Low)
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
18.2s
Cost:
$0.035610
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
98.51%
Dialogue to Total Word Ratio
(details)
99.2548%
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 unfolded the crayon map across the alley dumpster, weighting one corner with a half-eaten cupcake. Bill adjusted his black turtleneck, which squeaked because it was actually a wetsuit. "Security?" Molly whispered. "Three grandmothers, one magician, and a bouncy castle," Bill said. "The castle worries me." "It should. Inflatable fortifications are notoriously unstable." Bill tapped the drawing. "We enter through the petting zoo, disguise ourselves as goats, mingle, then grab the presents." "Goats don't carry sacks." "These goats have ambitious careers." Inside the community hall, children shrieked at musical chairs. Balloons bobbed like tiny informants. Molly peeked through the door and frowned. "New problem," she said. "Every present has a name tag." "So?" "So stealing Connor's spaceship from Connor feels personal." Bill sighed. "Fine. We steal only the untagged ones." "There aren't any." "Then we steal the cake." Molly watched a toddler hug a lumpy package as if it contained the moon. Her shoulders softened. "Bill, we're monsters." "We are professionals." "We're professional monsters." A magician produced a rabbit. The rabbit stared at Bill judgmentally. Bill zipped his sack closed, empty. "Retreat?" "Retreat," Molly said. "But take a balloon. For evidence." They fled, pursued only by music and shame.
created_at
"2026-04-24T21:09:21.212Z"
4/24/2026, 9:09:21 PM
model
"openai/gpt-5.5-20260423"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
15809
moderation_latency
(null)
generation_time
18046
tokens_prompt
77
tokens_completion
657
native_tokens_prompt
66
native_tokens_completion
1176
native_tokens_completion_images
0
native_tokens_reasoning
876
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
0.03561
router
(null)
provider_responses
0
endpoint_id
"58e5b336-423e-430b-a2ab-8bc353f0c51b"
id
"resp_05dc3f2b0715a8810169ebdc0152288190a1a8fd8d797dc6ec"
is_byok
false
latency
322
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-1777064961-Ia2RjasXovmh5tZdCjtD"
session_id
(null)
api_type
"completions"
id
"gen-1777064961-51nbg33Mi0YrOKn0kyfu"
upstream_id
"resp_05dc3f2b0715a8810169ebdc0152288190a1a8fd8d797dc6ec"
total_cost
0.03561
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
199
98.51%
Dialogue to Total Word Ratio
Ratio: 46.50%, Deviation: 3.50%
neededClean
false
wordsTotal
200
wordsDialogue
93
99.2548%