NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-03-18-01KM07JN5PESJ3VRBWVRP9EDHN
Run Details
Model:
GPT-5.4 Nano (Reasoning)
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
24.6s
Cost:
$0.003089
Provider:
OpenAI
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
99.86%
Dialogue to Total Word Ratio
(details)
99.9287%
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: “Tonight we steal every present like we’re allergic to joy.” Bill: “Joy is overrated. Also, I packed two pairs of socks—stolen socks.” Molly: “Disguise check. Your mustache says ‘mildly haunted librarian.’” Bill: “Your hoodie says ‘marshmallow fell into a craft store.’ Perfect.” Molly: “We enter during the group photo.” Bill: “Right. While the kids shout, we grab boxes and vanish.” Molly: “Do we take the tiny ones first?” Bill: “Yes. Small gifts are the stealthiest. They scream less.” Molly: “I heard the host’s mom is decorating with motion lights.” Bill: “Then we move like ninjas… with flash photography anxiety.” Molly: “No sticky hands this time.” Bill: “No frosting, no crumbs, no heroic cupcakes.” Molly: “If a child asks what we’re doing—” Bill: “We compliment their dinosaur helmet and pretend it’s a tour.” Molly: “Pretend we’re the birthday security.” Bill: “Exactly. ‘Please enjoy your loot responsibly.’” Molly: “What about the game prizes?” Bill: “We take only the ones labeled ‘for winners.’ Everyone here is a winner.” Molly: “On three?” Bill: “On three.” Molly: “One, two—” Bill: “—and on three, we become legends.” Molly: “And if we get caught?” Bill: “Then we blame the balloons. They’re too innocent for absolutely nothing.”
created_at
"2026-03-18T10:24:04.803Z"
3/18/2026, 10:24:04 AM
model
"openai/gpt-5.4-nano-20260317"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
24111
moderation_latency
(null)
generation_time
24407
tokens_prompt
77
tokens_completion
1446
native_tokens_prompt
66
native_tokens_completion
2461
native_tokens_completion_images
0
native_tokens_reasoning
2138
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
0.00308945
router
(null)
provider_responses
0
endpoint_id
"0c835f2e-c18d-4e8c-b245-e1e3bd08b97f"
id
"resp_0ac01eaf9155fccc0169ba7d44dd348193a46adbbfc7df97ec"
is_byok
false
latency
182
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-1773829444-AWRMim5PhjgXxudtHc3Z"
upstream_id
"resp_0ac01eaf9155fccc0169ba7d44dd348193a46adbbfc7df97ec"
total_cost
0.00308945
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
99.86%
Dialogue to Total Word Ratio
Ratio: 88.06%, Deviation: 1.94%
neededClean
false
wordsTotal
201
wordsDialogue
177
99.9287%