NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-16-01KHKAHPARTZD6YR0WFAPXM34B
Run Details
Model:
GPT-5 Mini
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
34.6s
Cost:
$0.005960
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: Window or party entrance? Bill: Gifts, cake, chaos. We do the cake. Molly: Kids scream, parents herd like confused sheep. Timing? Bill: PiƱata distraction, I bring portable kazoo. Might work. Molly: You're bringing a kazoo? Bill: It screams 'fun' and covers our exit. Also glitter bombs? Molly: Glitter equals chaos equals sticky evidence. No glitter. Bill: Fine. Balloon avalanche then. You good with helium? Molly: I'm good with helium. I'm not good with toddlers. Bill: I'll handle toddlers. I studied fast toddlers in camp. Molly: Put on your charming smile. Mine is on loan today. Bill: Charming smile is reserved for bankers. For parties I have a rubber duck puppet. Molly: Duck puppet might distract, but will it open gifts? Bill: I trained it. It prefers ribbon. It will fetch bows like a bloodhound. Molly: How do you teach a duck to unzip tape? Bill: Positive reinforcement and misplaced crackers. Also small letter opener. Molly: That's practical. What's our escape route? Bill: Back lawn, under the sprinkler. Kids hate mud. We run clean. Molly: We need bags. Camouflage: party bags. Pretense: gift inspectors. Bill: I'll wear a badge that says 'Certified Balloon Inflator.' Molly: Steal all the presents, then brunch?
created_at
"2026-02-16T13:35:02.890889+00:00"
2/16/2026, 1:35:02 PM
model
"openai/gpt-5-mini-2025-08-07"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
8283
moderation_latency
(null)
generation_time
34331
tokens_prompt
77
tokens_completion
1064
native_tokens_prompt
66
native_tokens_completion
2972
native_tokens_completion_images
0
native_tokens_reasoning
2624
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
0.0059605
router
(null)
provider_responses
0
id
"resp_02d36d0c1f04182e0169931ce3c1e88197b54ceaf656fc95db"
status
200
is_byok
false
latency
127
endpoint_id
"c4f66d01-20b0-4c27-a225-438ea22fda43"
provider_name
"OpenAI"
model_permaslug
"openai/gpt-5-mini-2025-08-07"
api_type
"completions"
id
"gen-1771248867-3dF0T1ptkUrUmYTi68YW"
upstream_id
"resp_02d36d0c1f04182e0169931ce3c1e88197b54ceaf656fc95db"
total_cost
0.0059605
cache_discount
(null)
upstream_inference_cost
0
provider_name
"OpenAI"
Evaluation details
Result
Evaluator
Details
Meta Data
99.9990%
Matches word count
n/a
neededClean
false
words
201
0.0000%
Dialogue to Total Word Ratio
Ratio: 3.98%, Deviation: 86.02%
neededClean
false
wordsTotal
201
wordsDialogue
8
49.9995%