NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-05-10-01KR92SYD749W8PG7H4S5V2Q2F
Run Details
Model:
Gemma 4 26B
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
8.8s
Cost:
$0.000121
Provider:
SiliconFlow
Evaluation summary
Score
Evaluator
Details Link
27.17%
Matches word count
(details)
99.67%
Dialogue to Total Word Ratio
(details)
63.4197%
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
"The target is the Miller residence, Bill. High-value toys, high-density sugar, and maximum chaos," Molly whispered, tapping a finger on the blueprint. "A children's birthday party? That's low, even for us," Bill grunted, adjusting his tactical turtleneck. "It’s not low, it’s tactical. The bouncy castle provides perfect acoustic cover for our entry." "And the smoke bombs?" "Too much drama. We use the glitter spray. It confuses the witnesses and masks our scent from the golden retriever." "What about the security? The birthday boy is eight. He's a menace." "He's a distraction, Bill. While he's busy blowing out candles, we hit the gift table. It’s a goldmine of Lego sets and remote-controlled drones." "And the loot bags? The ones with the tiny plastic whistles?" "Bag them immediately. They’re easy to flip on the black market." "I feel dirty, Molly. Really dirty." "Focus! Do you want the Lego Star Destroyer or not?" "Fine. But if a toddler asks me to play tag, I’m calling dibs on the getaway car." "Just keep your eyes on the cake, Bill. We move at two o'clock sharp."
created_at
"2026-05-10T13:56:37.937Z"
5/10/2026, 1:56:37 PM
model
"google/gemma-4-26b-a4b-it-20260403"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
4037
moderation_latency
(null)
generation_time
8677
tokens_prompt
77
tokens_completion
295
native_tokens_prompt
79
native_tokens_completion
280
native_tokens_completion_images
(null)
native_tokens_reasoning
0
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
0.00012148
router
(null)
provider_responses
0
endpoint_id
"eb382d64-dce1-42d5-b23d-6fefb91948a3"
id
"chatcmpl-aacc41711ac619cd"
is_byok
false
latency
4035
model_permaslug
"google/gemma-4-26b-a4b-it-20260403"
provider_name
"SiliconFlow"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.14.1; linux; x64))"
http_referer
(null)
request_id
"req-1778421397-SHzMbLbenGpnOKQw7Mlp"
session_id
(null)
api_type
"completions"
id
"gen-1778421397-5wfj8zMwbqunQFRhOZzQ"
upstream_id
"chatcmpl-aacc41711ac619cd"
total_cost
0.00012148
cache_discount
(null)
upstream_inference_cost
0
provider_name
"SiliconFlow"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
27.17%
Matches word count
n/a
neededClean
false
words
181
99.67%
Dialogue to Total Word Ratio
Ratio: 92.39%, Deviation: 2.39%
neededClean
false
wordsTotal
184
wordsDialogue
170
63.4197%