NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-10-13-01M4TAXMYCN46AGEN29S29BS6C
Run Details
Model:
Ember-1 (Reasoning, Max)
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
3m 14s
Cost:
$0.170222
Provider:
Fireworks
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
53.11%
Dialogue to Total Word Ratio
(details)
76.5567%
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
Molly unrolled the blueprint across the hood of the Buick, pinning one corner with a juice box she'd stolen for authenticity. The blueprint was drawn in crayon on the back of a placemat, because Molly believed presentation mattered. It was midnight in the Pancake Barn parking lot, which she considered the ideal venue for felony planning: unlimited coffee, and a waitress who minded her own business. "Right," she said. "Target: 42 Maple Lane. Saturday, two o'clock." "A kid's birthday party," Bill said flatly. "A kid's birthday party," Molly confirmed. "We've knocked over three banks, Molly." "And how much did we clear from those banks?" Bill scratched his chin and did the math on a napkin. "Negative four hundred, after the dye pack and the parking tickets." "Exactly. Banks are a scam. But this—" she tapped a crayon bouncy castle, "—this is a goldmine. The Henderson kid turns seven. Do you know what seven-year-olds get in this neighborhood?" "Nits?" "Presents. Tablets, consoles, a drone if the grandma's feeling guilty. Two grand of merchandise in a pile, unguarded." "Except for the parents." "Parents are the easiest mark alive. Exhausted, hopped up on frosting, distracted by thirty screaming children. We go in as the entertainment." Bill squinted at her. "What entertainment?" Molly reached into the back seat and produced two clown wigs. Bill looked at them the way a man looks at a diagnosis. "Absolutely not." "You get to be Bongo." "I want to be Dazzle." "Dazzle's mine. I planned the job." Bill took the Bongo wig with the resignation of a man who had stopped fighting for his dignity around the time of the dye pack. He held it up to the parking lot lights. It smelled faintly of a previous owner's poor decisions, which, in fairness, so did Bill. "Walk me through it," he sighed. Molly's finger traced the crayon route, past a stick-figure dog she'd drawn for morale. "We arrive at one-forty-five. I do balloon animals to build trust—" "You can't do balloon animals." "I can do a snake." "That's just a balloon." "A snake is the gateway animal. While I'm dazzling, you hit the gift table. Pillowcase, ninety seconds. We exit during the piñata—nobody looks anywhere but at a piñata." Bill considered this, stirring his coffee with the wig's tag still dangling from his fist. "What if the kids see us?" "Kids are terrible witnesses." "What if one bites?" "Then you've learned something about yourself." He turned the wig over. A moth flew out, took one look at the situation, and left. "Molly. Be honest. Is this a new low?" Molly folded the blueprint with great care, creasing it along the bouncy castle. "Bill, last month we stole commemorative spoons from a Denny's. This is a step up. This is a score with a theme." "A seven-year-old's theme." "Dinosaurs," Molly said, offended. "The theme is dinosaurs." Bill climbed into the passenger seat, Bongo's wig already on his head, because some battles were not worth fighting twice.
created_at
"2026-10-13T09:53:55.159Z"
10/13/2026, 9:53:55 AM
model
"fireworks/ember-1-20260923"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
2457
moderation_latency
(null)
generation_time
194008
tokens_prompt
75
tokens_completion
10385
native_tokens_prompt
149
native_tokens_completion
11345
native_tokens_completion_images
(null)
native_tokens_reasoning
10385
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
0.1702224
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.21.0; linux; x64))"
http_referer
(null)
request_id
"req-1791885235-cTbfIb3bCiywJu37s6Kv"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1791885235-RYUslZzvQMujDmKpSdCL"
upstream_id
"chatcmpl-4c8ba12ba1c84e29b0556ac0eb411108"
provider_responses
0
endpoint_id
"b89a2f05-4051-4446-a9b0-7bc8ae7b7c90"
id
"chatcmpl-4c8ba12ba1c84e29b0556ac0eb411108"
is_byok
false
latency
2457
model_permaslug
"fireworks/ember-1-20260923"
provider_name
"Fireworks"
status
200
total_cost
0.1702224
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
100.00%
Matches word count
n/a
neededClean
false
words
499
53.11%
Dialogue to Total Word Ratio
Ratio: 46.44%, Deviation: 3.56%
neededClean
false
wordsTotal
506
wordsDialogue
235
76.5567%