NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-16-01KHK9NRQJ6ZZQAXJRWHF07QKQ
Run Details
Model:
GPT-5 Mini
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
30.4s
Cost:
$0.006306
Provider:
OpenAI
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: We steal the presents tomorrow, we get rich and the children cry. Bill: The children crying is a feature, not a bug. Molly: No alarms, no sneaky tools, just style and misdirection. Bill: Misdirection? Last time you misdirected us into a charity bake sale. Molly: That cake bought us new boots. Bill: Boots don't fit if the heel collapses during a getaway. Molly: We wear sensible footwear and sing terrible jingles. Bill: Singing? The neighbors will call for therapy. Molly: Therapy is expensive; stealing is cheaper. Bill: We're thieves with hearts of coal and pockets of lint. Molly: Lint can hide treasure. Also glitter. Bill: Glitter betrays us forever. Molly: We'll use confetti camouflage, then vanish into a piñata storm. Bill: Piñatas are political now. Molly: Fine, biodegradable illusions. Bill: Do we have a plan or just optimism and glitter? Molly: Optimism, glitter, and a borrowed mascot costume. Bill: Borrowed from whom? Molly: Someone very trusting. Bill: I prefer a plan that doesn't involve costumes named Larry. Molly: Larry is committed to the bit. Bill: Then commit to leaving at dawn. Molly: Dawn is dramatic. Night is theatrical. Bill: Night is suspicious. Molly: Suspicion is our signature. Bill: Agreed.
created_at
"2026-02-16T13:19:43.28565+00:00"
2/16/2026, 1:19:43 PM
model
"openai/gpt-5-mini-2025-08-07"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
6522
moderation_latency
(null)
generation_time
30251
tokens_prompt
77
tokens_completion
924
native_tokens_prompt
66
native_tokens_completion
3145
native_tokens_completion_images
0
native_tokens_reasoning
2816
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
0.0063065
router
(null)
provider_responses
0
id
"resp_0f411769dffca81c0169931950bf8c8195ba140d90405e35af"
status
200
is_byok
false
latency
159
endpoint_id
"c4f66d01-20b0-4c27-a225-438ea22fda43"
provider_name
"OpenAI"
model_permaslug
"openai/gpt-5-mini-2025-08-07"
api_type
"completions"
id
"gen-1771247952-wyoymGQX2qeOzsgEUdyu"
upstream_id
"resp_0f411769dffca81c0169931950bf8c8195ba140d90405e35af"
total_cost
0.0063065
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
Evaluation details
Result
Evaluator
Details
Meta Data
100.0000%
Matches word count
n/a
neededClean
false
words
200
0.0000%
Dialogue to Total Word Ratio
Ratio: 0.00%, Deviation: 90.00%
neededClean
false
wordsTotal
200
wordsDialogue
0
50.0000%