NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-10-13-01M4TBPD2ET3ET0MA1SWFQ34T7
Run Details
Model:
Ember-1 (Reasoning, Max)
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
3m 46s
Cost:
$0.215207
Provider:
Fireworks
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
75.93%
Dialogue to Total Word Ratio
(details)
37.9665%
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
The van idled across the street from a house wearing balloons on its mailbox like a warning. Molly wore black, as always, though the only thing black attracted at a children's party was suspicion and frosting. She unrolled a blueprint, hand-drawn in crayon and stolen from the birthday boy's older sister, across the hood. "Target's a split-level on Maple. Party at two. We hit at two-forty-five." Bill squinted through a mouthful of gas-station sandwich. "Why two-forty-five?" "Musical chairs. Peak chaos. Nobody watches the gift table." Bill had that sinking feeling again. She caught his look. "I've thought about nothing else for three weeks." She tapped the dining room, where, through a distant window, a pyramid of wrapped boxes glittered like the world's least secure vault. "Presents pile up here. Linda the mom runs a tight ship, but crumbles if the magician's late." "There's a magician?" "There was a magician. His car has a flat on Route 9." Bill lowered the sandwich. "You gave a magician a flat tire." "He'll survive. He's a magician." From the backyard came the shriek of children discovering cake. Molly checked her watch, calm as a surgeon. Bill was less calm. "PiƱata at three. Six minutes while they whack a donkey. Twenty-two RSVPs, two generous grandparents, one rich uncle. Call it thirty boxes." "Of what, though?" Molly hesitated. "Toys." "Toys." "And gift cards. Kids love gift cards." Bill rubbed his temples, leaving a smear of mustard he didn't notice. "We robbed a jewelry store in March. We had a tunnel." "The tunnel flooded." "We had a trajectory. Upward." "It's low-risk. Nobody expects a heist at a six-year-old's birthday." "Because it's insane!" "Because it's genius. No alarms, no cameras. Just Linda, twenty-two sticky kids, and Kevin the clown." "Kevin worries me." "Kevin's afraid of balloons and works with balloons. Kevin has problems." As if summoned, Kevin pedaled past on a bicycle, wig askew, radiating the despair of a man who had made choices. Neither thief acknowledged him. Bill folded his arms. "What's the take?" "Eight hundred in gift cards, a drone, and a Lego set retailing at two-fifty." "The castle one?" "The castle one." A long pause. Somewhere, a party blower unfurled. "That's a good set." "Right? These kids get so much stuff they won't miss six boxes. Remember your sixth birthday?" "No." "Exactly. No lasting damage." Bill sighed, then folded the blueprint with the care of a man accepting his fate. "Where do I park?" "Behind the bounce house. Sight lines are dead there." He gestured helplessly at himself. Molly grinned and tossed him a bag of long, skinny balloons. "You're the balloon artist." "I can't make balloon animals." "You can make one thing: a snake. It's a balloon you don't do anything to." Bill turned a balloon over like it might contain deeper meaning. "That's actually true." "See? Born for this job." "I was born for better than this job." "We all were. Get in the van." He got in the van. Across the street, twenty-two children were eating cake, blissfully unaware that two professionals were coming for their plastic treasures. Bill practiced his balloon snake the whole way around the block. It was perfect every time. It was also, he reflected, just a balloon.
created_at
"2026-10-13T10:07:26.295Z"
10/13/2026, 10:07:26 AM
model
"fireworks/ember-1-20260923"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
491
moderation_latency
(null)
generation_time
226246
tokens_prompt
75
tokens_completion
13230
native_tokens_prompt
149
native_tokens_completion
14344
native_tokens_completion_images
(null)
native_tokens_reasoning
13230
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
0.2152074
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.21.0; linux; x64))"
http_referer
(null)
request_id
"req-1791886046-zCHFsZE9XR2dFt07Y14O"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1791886046-wgR5o5gWNQ9ySDLh8d2B"
upstream_id
"chatcmpl-1a67c560e30a4923885454dd93273986"
provider_responses
0
endpoint_id
"b89a2f05-4051-4446-a9b0-7bc8ae7b7c90"
id
"chatcmpl-1a67c560e30a4923885454dd93273986"
is_byok
false
latency
491
model_permaslug
"fireworks/ember-1-20260923"
provider_name
"Fireworks"
status
200
total_cost
0.2152074
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
0.00%
Matches word count
n/a
neededClean
false
words
540
75.93%
Dialogue to Total Word Ratio
Ratio: 52.35%, Deviation: 2.35%
neededClean
false
wordsTotal
554
wordsDialogue
290
37.9665%