NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-10-13-01M4TB6592B68J0MF7K97XV3NQ
Run Details
Model:
Ember-1 (Reasoning, Max)
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
1m 15s
Cost:
$0.067577
Provider:
Fireworks
Evaluation summary
Score
Evaluator
Details Link
27.17%
Matches word count
(details)
93.61%
Dialogue to Total Word Ratio
(details)
60.3867%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 500 word scene that has 50% 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
Bill spread the map across the hood of the Buick, anchoring the corners with a half-eaten burrito and one of Molly's socks. "That's a coloring book page," Molly said. "It's the community center. Timmy's mom posted photos online. I just added the bouncy castle." Molly squinted at the crayon lines. "Why is the bouncy castle labeled 'danger zone'?" "Have you ever been inside one? They're lawless. No one comes out the same." Molly opened her duffel bag to reveal a clown wig, two water pistols, and a beige cardigan. "The cardigan makes us look like parents," she said. "You look like a substitute teacher who cries in her car." She held the cardigan against her shoulders and checked her reflection in the old Buick's side mirror. "Perfect. Parents trust her. That's the whole job, Bill. Look trustworthy and carry a big bag." Bill tapped the map with a glitter pen. The plan was simple. They'd arrive at two, during the magician, when every eye in the room would be glued to a man pulling a rabbit out of a hat. Molly would create a diversion near the cake table. Bill would sweep the gift table into a garbage bag disguised as a gift bag, which he had labeled in advance for authenticity. He had practiced the sweep in his garage with a folding table and a stack of phone books, and it had gone beautifully except for the noise. "Wait," Molly said. "Read what your bag says." "It says 'presents.' So nobody questions it." "It says 'STOLEN presents.'" Bill rotated the bag slowly. "Huh. The 'stolen' was supposed to go on the inside." "Your penmanship is why we can't have nice things." "We're thieves. We're not supposed to have nice things. That's the whole point of the profession." Molly crossed her arms and looked at the map, then at the bag, then at the horizon, where somewhere a six-year-old was probably practicing his happy birthday face in a mirror. "These are kids, Bill. Little kids. With little faces." "Little faces that don't need forty presents. We're doing them a favor. It builds character." "What character does it build when your Nerf gun gets stolen?" "Resilience. Grit. Timmy will thank us at his graduation." "Timmy's dad is a cop." Bill went quiet for a moment, folding the map very slowly, the way a man folds when recalculating his entire life. "Okay," he said. "New plan. Same plan, but faster, and we don't touch anything shaped like a badge." "And the cardigan stays," Molly said, pulling it on. "The cardigan stays. But if the magician sees through us, I'm not saving you." "He's a birthday magician, Bill. He once pulled a rabbit out of a hat." "Exactly. If he can do that, God knows what else he can pull out of there." They got in the Buick. Molly adjusted the wig in the mirror. Bill put the car in drive, then paused. "We're good people," he said. "We're stealing birthday presents from a child." "Good people, bad errands." He started the engine. It coughed, considered, and finally committed to the crime.
created_at
"2026-10-13T09:58:34.027Z"
10/13/2026, 9:58:34 AM
model
"fireworks/ember-1-20260923"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
515
moderation_latency
(null)
generation_time
74527
tokens_prompt
75
tokens_completion
4518
native_tokens_prompt
149
native_tokens_completion
4502
native_tokens_completion_images
(null)
native_tokens_reasoning
3797
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
0.0675774
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.21.0; linux; x64))"
http_referer
(null)
request_id
"req-1791885514-hA2SncpKyMSY3ixIBmk1"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1791885514-4lo6eHcPoEmWFDplRzFL"
upstream_id
"chatcmpl-b51923b2f4324e7e9a882e92041f9a82"
provider_responses
0
endpoint_id
"b89a2f05-4051-4446-a9b0-7bc8ae7b7c90"
id
"chatcmpl-b51923b2f4324e7e9a882e92041f9a82"
is_byok
false
latency
515
model_permaslug
"fireworks/ember-1-20260923"
provider_name
"Fireworks"
status
200
total_cost
0.0675774
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
27.17%
Matches word count
n/a
neededClean
false
words
519
93.61%
Dialogue to Total Word Ratio
Ratio: 48.85%, Deviation: 1.15%
neededClean
false
wordsTotal
522
wordsDialogue
255
60.3867%