NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-10-13-01M4TBMZYCQX34V8FCMPHAC8T4
Run Details
Model:
Ember-1 (Reasoning, Max)
Test:
Dialogue tags
Scenario:
Write 500 words with 30% dialogue
Duration:
2m 34s
Cost:
$0.151967
Provider:
Fireworks
Evaluation summary
Score
Evaluator
Details Link
90.48%
Matches word count
(details)
53.53%
Dialogue to Total Word Ratio
(details)
72.0066%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 500 word scene that has 30% 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 unrolled the blueprint across the hood of the car, pinning the corners with a juice box and half a granola bar. Bill recognized the Henderson place at once. He had driven past it twice that week, slowing each time to admire the mountain of wrapped gifts visible through the front window. Molly had been watching it for three days, which she insisted was standard practice for a job of this magnitude. "Saturday, two to five," Molly said. "One birthday boy, age seven. Twenty guests. Estimated haul: thirty presents, minimum." Bill squinted at the diagram. She had drawn the bouncy castle to scale, complete with a tiny stick figure mid-bounce. She had mapped every adult in the yard too: a father chained to the grill, a mother running party games, and one clown. Bill felt his stomach tighten. "A clown," Bill said. "Balloon animals. Goes by Mr. Wiggles. I couldn't get a read on him." "Clowns are unpredictable." "Exactly why you're the distraction." "I'm not dressing as a clown." "You're dressing as a magician. There's a difference." "What's the difference?" "A magician has a hat." The plan itself was elegant. At three o'clock the cake would emerge, and every adult head would swivel toward the patio like sunflowers. Bill would be stationed at the gift table performing card tricks for the children. "Kids are terrible witnesses," Molly said. "One of them once identified a mailbox in a lineup." While Bill dazzled the ankle-biters with the disappearing thumb, a trick that had cost him six weeks and a friendship, Molly would slip through the sliding door, load the presents into a laundry hamper, and exit through the side gate before the candles were blown. "That's a ninety-second window," Bill said. "Tight." "You stole a catalytic converter in forty." "That was different. That was mine." There were complications, naturally. One rectangular present, wrapped in dinosaur paper, was suspiciously heavy. "Who gives a kid a brick?" Bill asked. "Uncles, Bill. Uncles are the wild card of every operation." Worse was the piñata, hanging directly above the escape route like a papier-mâché sword of Damocles stuffed with fun-size chaos. Molly had circled it in red, then circled the circle. "If the piñata breaks early, we have a candy situation," Molly said. "Candy everywhere. Kids screaming. The one variable I can't control." "So what's the contingency?" "Grab what's in reach and run." Bill nodded slowly. It was honest, at least. The best plans always came down to that eventually. They stood a while longer, regarding the blueprint with the gravity of two professionals whose combined career earnings totaled nine hundred dollars and a breadstick. "One more thing," she said. "If you see a goodie bag, take it. I hear they've got those little rubber duckies this year." Bill smiled for the first time all afternoon. Somewhere behind them a lawnmower started up, and the neighborhood went on ignoring them completely. "Now you're speaking my language."
created_at
"2026-10-13T10:06:40.086Z"
10/13/2026, 10:06:40 AM
model
"fireworks/ember-1-20260923"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
498
moderation_latency
(null)
generation_time
153490
tokens_prompt
75
tokens_completion
9614
native_tokens_prompt
149
native_tokens_completion
10128
native_tokens_completion_images
(null)
native_tokens_reasoning
9467
native_tokens_cached
148
num_media_prompt
(null)
num_input_audio_prompt
(null)
num_media_completion
0
num_search_results
(null)
num_fetches
(null)
web_search_engine
(null)
origin
"https://ncbench.com/"
is_byok
false
finish_reason
"stop"
native_finish_reason
"stop"
service_tier
(null)
usage
0.1519674
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.21.0; linux; x64))"
http_referer
(null)
request_id
"req-1791886000-N7ekOx0FDEEZ3OkND8cC"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1791886000-iZIE4FL0rJv204MNDnGj"
upstream_id
"chatcmpl-0df4784b78c64cb2a20c3c5c5fcf7a7e"
provider_responses
0
endpoint_id
"b89a2f05-4051-4446-a9b0-7bc8ae7b7c90"
id
"chatcmpl-0df4784b78c64cb2a20c3c5c5fcf7a7e"
is_byok
false
latency
498
model_permaslug
"fireworks/ember-1-20260923"
provider_name
"Fireworks"
status
200
total_cost
0.1519674
cache_discount
0.0003996
upstream_inference_cost
0
provider_name
"Fireworks"
response_cache_source_id
(null)
data_region
"global"
workspace_id
"97e315e5-d303-487d-83c1-83180e8a13d4"
Evaluation details
Result
Evaluator
Details
Meta Data
90.48%
Matches word count
n/a
neededClean
false
words
490
53.53%
Dialogue to Total Word Ratio
Ratio: 33.54%, Deviation: 3.54%
neededClean
false
wordsTotal
495
wordsDialogue
166
72.0066%