NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-18-01KM07BW8N2CKZ5F4DMA4YMDV3
Run Details
Model:
GPT-5.4 Nano (Reasoning, Low)
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
10.1s
Cost:
$0.002653
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
0.00%
Dialogue to Total Word Ratio
(details)
49.9995%
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: Bill, did you bring the tiny crowbar? Bill: Yes—two, one for prying, one for confidence. Molly: Perfect. The party’s at the community center at six sharp. Bill: Sharp like “spilled soda.” Molly: Exactly. We get in, we steal the presents, we vanish before someone asks questions. Bill: I practiced my smile. It says, “I’m friendly.” Molly: Your smile says, “I’m a raccoon in pants.” Bill: Raccoons respect teamwork. Molly: We wear the costume bins. Bill: The clown suits? Molly: Technically “space explorers.” Bill: Space explorers steal? Molly: Only on our mission. Bill: How do we bypass the door guard? Molly: The mascot costume. Bill: We’re not the mascot. Molly: We’ll become it. Bill: Like I’m not a toaster, I’m just near bread. Molly: Don’t talk. See Dad with the clipboard? Bill: The one counting invitations. Molly: We borrow it, then stamp “Merry Schemingsmas.” Bill: That’s not a holiday. Molly: It is tonight. Molly: After the presents, we swap the tags. Bill: For fake ones? Molly: For identical ones, but wrapped in “wrong kid” paper. Bill: That’s evil. Molly: It’s efficient. Bill: And the toys? Molly: Emergency strategy: steal batteries too. Bill: Great, I’ll carry the drill like a flashlight.
created_at
"2026-03-18T10:20:22.689Z"
3/18/2026, 10:20:22 AM
model
"openai/gpt-5.4-nano-20260317"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
8994
moderation_latency
(null)
generation_time
9922
tokens_prompt
77
tokens_completion
324
native_tokens_prompt
66
native_tokens_completion
2112
native_tokens_completion_images
0
native_tokens_reasoning
1794
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
0.0026532
router
(null)
provider_responses
0
endpoint_id
"0c835f2e-c18d-4e8c-b245-e1e3bd08b97f"
id
"resp_012e4b85a8337a3e0169ba7c66c5408197ab2f7f6e8b4153da"
is_byok
false
latency
243
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-1773829222-CXY3eE2VXuWfSnHG1zto"
upstream_id
"resp_012e4b85a8337a3e0169ba7c66c5408197ab2f7f6e8b4153da"
total_cost
0.0026532
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
Evaluation details
Result
Evaluator
Details
Meta Data
100.00%
Matches word count
n/a
neededClean
false
words
201
0.00%
Dialogue to Total Word Ratio
Ratio: 7.46%, Deviation: 82.54%
neededClean
false
wordsTotal
201
wordsDialogue
15
49.9995%