NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-11-19-01M7SVSD6JGPKMKCP3DX4FDKAK
Run Details
Model:
Gemini 4 Hydrogen (Reasoning, Low)
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
11.2s
Cost:
$0.012948
Provider:
Google
Evaluation summary
Score
Evaluator
Details Link
68.10%
Matches word count
(details)
99.96%
Dialogue to Total Word Ratio
(details)
84.0295%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 90% 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 tapped the blueprint. "Okay, Bill, the target is the Henderson backyard. Seven-year-old's birthday. We go in at three." "Three? That's peak sugar-rush hour, Molly. Those kids will be feral." "Precisely. While they're distracted by the magician, we hit the gift table. Grab the Lego sets, leave the socks." "What about the bouncy castle? It's blocking the side gate." "We deflate it. Quietly." "With a pin? It'll sound like a gunshot, and thirty toddlers will stampede us." "Fine. We crawl under it. You're good at tight spaces." Bill sighed, rubbing his temples. "I'm forty-two, Molly, my knees sound like bubble wrap." "Then bring knee pads. Look, the haul is worth at least four hundred bucks in plastic." "Four hundred? We're risking juice-box stains and sticky fingers for four hundred bucks?" "Plus, I heard there's a custom-built PlayStation cake." "You're stealing the cake too? That's cold." "I'm a thief, Bill, not a monster. I'll leave them a cupcake." "Alright, I'm in. Should I wear the clown mask?" "No! Last time you wore that, we had to bribe a six-year-old with my watch just to stop crying." "Fair point. Ski masks it is." Molly rolled her eyes and tossed him a black beanie. "Just don't eat the frosting before we're in the van." "No promises, Molly."
created_at
"2026-11-19T12:16:32.985Z"
11/19/2026, 12:16:32 PM
model
"google/gemini-4-hydrogen-20261112"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1525
moderation_latency
(null)
generation_time
11187
tokens_prompt
75
tokens_completion
684
native_tokens_prompt
97
native_tokens_completion
1707
native_tokens_completion_images
0
native_tokens_reasoning
1347
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
"default"
usage
0.012948
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.22.2; linux; x64))"
http_referer
(null)
request_id
"req-1795090592-ngGSmmw1Obet5SsuilpL"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1795090592-vflBCvE8rgHLblRFJ2zk"
upstream_id
"oej-aoaxAuLvsbwPv4CPqAg"
provider_responses
0
endpoint_id
"3936f89d-7d96-43b5-adaf-f690b9de1298"
id
"oej-aoaxAuLvsbwPv4CPqAg"
is_byok
false
latency
1525
model_permaslug
"google/gemini-4-hydrogen-20261112"
provider_name
"Google"
status
200
total_cost
0.012948
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Google"
response_cache_source_id
(null)
data_region
"global"
workspace_id
"97e315e5-d303-487d-83c1-83180e8a13d4"
Evaluation details
Result
Evaluator
Details
Meta Data
68.10%
Matches word count
n/a
neededClean
false
words
214
99.96%
Dialogue to Total Word Ratio
Ratio: 91.44%, Deviation: 1.44%
neededClean
false
wordsTotal
222
wordsDialogue
203
84.0295%