NC Bench
Overview
About
Tests
Models
Vendors
Run: 2026-02-18-01KHQ92RC39EM4V52MQRE2DBS5
Run Details
Model:
Claude Sonnet 4.6
Test:
Dialogue tags
Scenario:
Write 200 words with 50% dialogue
Duration:
8.0s
Cost:
$0.004800
Provider:
Anthropic
Evaluation summary
Score
Evaluator
Details Link
93.65%
Matches word count
(details)
0.02%
Dialogue to Total Word Ratio
(details)
46.8362%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 200 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 hand-drawn map across the hood of their rusty van. She had labeled the bounce house "Threat Level: Moderate" and the pony ride "Do Not Engage." "The presents are stacked near the gazebo," she whispered, tapping the map with her finger. "We go in during Pin the Tail on the Donkey. Maximum chaos, minimum visibility." Bill squinted at the map. "You drew the clown twice." "Because there are two clowns, Bill. I did reconnaissance." "You went to a seven year old's birthday party alone?" "I wore a disguise." Bill stared at her. "What disguise?" "A smaller hat." He rubbed his temples slowly. "Molly, we are adults trying to steal presents from children." "Allegedly steal. And yes. Focus." "I feel like we should talk about our life choices." "Talk later, presents now." She rolled up the map with authority. "You handle the getaway. I handle the gift table. Nobody touches the piƱata, that's a distraction for the parents." Bill looked genuinely impressed despite himself. "You really thought this through." "I am a professional." "You are absolutely not a professional." "Get in the van, Bill." He got in the van.
created_at
"2026-02-18T02:25:55.811243+00:00"
2/18/2026, 2:25:55 AM
model
"anthropic/claude-4.6-sonnet-20260217"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
453
moderation_latency
(null)
generation_time
7983
tokens_prompt
77
tokens_completion
312
native_tokens_prompt
75
native_tokens_completion
305
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)
origin
"https://ncbench.com/"
is_byok
false
finish_reason
"stop"
native_finish_reason
"stop"
usage
0.0048
router
(null)
provider_responses
0
id
"msg_01T2gJhbeee1e5tucHdcKHp9"
status
200
is_byok
false
latency
453
endpoint_id
"6d408764-7dd7-4626-bb87-a6cc1589bc86"
provider_name
"Anthropic"
model_permaslug
"anthropic/claude-4.6-sonnet-20260217"
api_type
"completions"
id
"gen-1771381547-tr8TYYZByg3wUgykASbr"
upstream_id
"msg_01T2gJhbeee1e5tucHdcKHp9"
total_cost
0.0048
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Anthropic"
Evaluation details
Result
Evaluator
Details
Meta Data
93.65%
Matches word count
n/a
neededClean
false
words
191
0.02%
Dialogue to Total Word Ratio
Ratio: 67.02%, Deviation: 17.02%
neededClean
false
wordsTotal
191
wordsDialogue
128
46.8362%