NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-05-10-01KR92TTPSB7XD31GGPMW3FSHK
Run Details
Model:
Gemma 4 26B (Reasoning)
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
6m
Cost:
$0.003629
Provider:
DeepInfra
Evaluation summary
Score
Evaluator
Details Link
0.00%
Matches word count
(details)
1.90%
Dialogue to Total Word Ratio
(details)
0.9492%
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 slammed the glitter-encrusted invitation onto the scarred wooden table. The pink sparkles drifted through the air like tiny, mocking diamonds, catching the dim, flickering light of the damp basement. "This is the target, Bill. Operation Sugar Rush is officially a go," Molly whispered. Bill leaned in, squinting through his thick, cracked spectacles that sat crooked on his nose. "Is that a dinosaur sticker on the envelope? Are we hitting a preschool?" "It is a seventh birthday, Bill. Do not be insulting," Molly snapped, pulling a heavy magnifying glass from her utility belt. "I just think we could be doing something bigger. Like a bank. Not a backyard with a clown," Bill said, picking a stubborn piece of lint off his oversized tactical vest. Molly tapped a finger on the map, which was actually a hand-drawn crayon diagram of a suburban cul-de-sac, complete with smiley faces. "The payout is massive. Think of the volume: Lego sets, Nintendo Switches, and rare cards." "What about security? I heard there is a dog named Buster who is not house-trained," Bill asked, his voice trembling slightly. "Buster is secondary. The real threat is the bouncy castle," Molly said, her voice dropping an octave. "The structural instability makes stealth nearly impossible. One wrong step and we are bouncing into the neighbors' yard." Bill leaned back, the old wooden chair creaking precariously under his weight. "Can we wait until the cake is served? The sugar high is a perfect diversion." "That is what they want. We strike the gift pile while they are vibrating with glucose," Molly replied, pacing the small room. "We enter through the hedge and secure the goods before the piƱata is cracked." Bill sighed, rubbing his temples with weary fingers. "What if a kid sees us? I am not getting tackled by a Spider-Man again." "Maintain composure. You are a professional thief, not a playground regular," Molly commanded. "Professionals do not usually carry juice boxes in tactical gear," Bill muttered, glancing at the orange pouches on his belt. "It is for the diversion! We will be two serious uncles looking for the dip," Molly explained, gesturing wildly. Bill looked at the invitation again, his brow furrowed. "Do I have to wear a party hat?" "Only if it helps you blend in," Molly said, turning back to the map. "If I get hit with a water balloon, I am retiring," Bill grumbled. Molly nodded, her expression turning grim. "Prepare the van. Line the trunk with bubble wrap so nothing rattles." "Should I bring the extra napkins?" Bill asked. "Yes, Bill. Bring the napkins," Molly replied.
created_at
"2026-05-10T13:57:06.912Z"
5/10/2026, 1:57:06 PM
model
"google/gemma-4-26b-a4b-it-20260403"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
590
moderation_latency
(null)
generation_time
359648
tokens_prompt
77
tokens_completion
9985
native_tokens_prompt
82
native_tokens_completion
10657
native_tokens_completion_images
(null)
native_tokens_reasoning
9302
native_tokens_cached
0
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
0.00362912
router
(null)
provider_responses
0
endpoint_id
"d71c6981-347d-4622-a86b-2c9e5d0f2c6c"
id
"chatcmpl-RujOO15H5wTmdP5qW8kYdWzb"
is_byok
false
latency
427
model_permaslug
"google/gemma-4-26b-a4b-it-20260403"
provider_name
"DeepInfra"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.14.1; linux; x64))"
http_referer
(null)
request_id
"req-1778421426-R5CXGx5RQd73mEkXRmHj"
session_id
(null)
api_type
"completions"
id
"gen-1778421426-xJVVfGKf5wFo624gRKUb"
upstream_id
"chatcmpl-RujOO15H5wTmdP5qW8kYdWzb"
total_cost
0.00362912
cache_discount
(null)
upstream_inference_cost
0
provider_name
"DeepInfra"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
0.00%
Matches word count
n/a
neededClean
false
words
432
1.90%
Dialogue to Total Word Ratio
Ratio: 58.90%, Deviation: 8.90%
neededClean
false
wordsTotal
438
wordsDialogue
258
0.9492%