NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-18-01KM087HAG1XG02WK7K3ZG3DQV
Run Details
Model:
GPT-5.4 Nano (Reasoning, Low)
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
3.0s
Cost:
$0.000737
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
100.00%
Dialogue to Total Word Ratio
(details)
50.0000%
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: “Okay, Bill, whisper—this is a birthday party, not a courtroom.” Bill: “I’m whispering with confidence. Confidence is quiet. Stealing is quiet-ish.” Molly: “No ‘ish.’ We’re taking every present. Not opening them. Just relocating them to our bag of destiny.” Bill: “Bag of destiny sounds illegal. Can we call it ‘mildly suspicious duffel’?” Molly: “Fine. Mildly suspicious duffel. First move: disguise.” Bill: “I have the perfect disguise.” Molly: “Let me guess—your face taped to a clown costume again?” Bill: “It’s a learning process. The clown gets better, eventually.” Molly: “Your clown is unionized and demands better padding.” Bill: “Relax. I also brought a decoy. I’m wearing a hat that says ‘NOT FOR YOU’.” Molly: “That’s what criminals always wear.” Bill: “That’s what successful criminals always wear.” Molly: “Listen, the kids are distracted by cake. We strike during frosting chaos.” Bill: “Frosting chaos is my specialty. I can sense sprinkles like a bloodhound senses guilt.” Molly: “Focus. You go left; I go right. If someone asks questions, we’re ‘party helpers.’” Bill: “And if the party host asks for help?” Molly: “We’ll ‘help’ by dramatically holding a clipboard we can’t read.” Bill: “Great. I’ll drop confetti to cover our footsteps.” Molly: “Confetti covers footsteps like a napkin covers taxes.” Bill: “Taxes are heavier than presents.” Molly: “Exactly. Now, bags ready?” Bill: “Ready. I’m ready. Emotionally, I’m also ready.” Molly: “Silence, then go.” Bill: “I’m going—quietly—like a magician stealing applause.”
created_at
"2026-03-18T10:35:28.992Z"
3/18/2026, 10:35:28 AM
model
"openai/gpt-5.4-nano-20260317"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1276
moderation_latency
(null)
generation_time
2658
tokens_prompt
77
tokens_completion
472
native_tokens_prompt
66
native_tokens_completion
579
native_tokens_completion_images
0
native_tokens_reasoning
147
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
0.00073695
router
(null)
provider_responses
0
endpoint_id
"0c835f2e-c18d-4e8c-b245-e1e3bd08b97f"
id
"resp_0c2d21be2fec036e0169ba7ff11a148195a1b1638eb8ad6421"
is_byok
false
latency
238
model_permaslug
"openai/gpt-5.4-nano-20260317"
provider_name
"OpenAI"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.14.0; linux; x64))"
http_referer
(null)
api_type
"completions"
id
"gen-1773830128-HiFt6T5yvnIimZZFieR4"
upstream_id
"resp_0c2d21be2fec036e0169ba7ff11a148195a1b1638eb8ad6421"
total_cost
0.00073695
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
Evaluation details
Result
Evaluator
Details
Meta Data
0.00%
Matches word count
n/a
neededClean
false
words
241
100.00%
Dialogue to Total Word Ratio
Ratio: 90.08%, Deviation: 0.08%
neededClean
false
wordsTotal
242
wordsDialogue
218
50.0000%