NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-05-31-01KSYXCJHBYZXD275SPEXV51VG
Run Details
Model:
Claude Opus 4.8 (Reasoning, High)
Test:
Dialogue tags
Scenario:
Write 200 words with 10% dialogue
Duration:
11.0s
Cost:
$0.015065
Provider:
Anthropic
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
90.71%
Dialogue to Total Word Ratio
(details)
95.3542%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 10% 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 crouched behind the hydrangeas, binoculars trained on the bouncy castle wobbling in the suburban backyard. Streamers sagged from the gutters. A clown honked somewhere inside, sounding less festive than vaguely menacing. Beside her, Bill unrolled a hand-drawn map across his knees, smearing it with the cupcake frosting he'd stolen earlier as reconnaissance. "The presents are stacked by the piñata," he whispered. She nodded, scanning the perimeter. Two mothers guarded the snack table like prison wardens. A toddler armed with a juice box patrolled the gate, suspicious of everyone. Worst of all, a golden retriever named, according to the banner, Sir Barkington, dozed beside the gift mountain—wrapping paper glinting like treasure. Molly's plan was elegant. Bill would create a diversion by joining the limbo line, his criminal flexibility finally serving a purpose. Meanwhile she'd slip in disguised as the entertainment, having purchased a secondhand fairy costume that fit nobody comfortably. The wings alone could knock over a six-year-old. She checked her watch, then the cake, then her dignity, which had departed hours ago. "On three," she breathed. Bill swallowed nervously, adjusting his balloon-animal hat. Somewhere a child screamed about cake. Destiny called, frosted and ribboned, and the thieves answered.
created_at
"2026-05-31T11:40:56.249Z"
5/31/2026, 11:40:56 AM
model
"anthropic/claude-4.8-opus-20260528"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
3437
moderation_latency
(null)
generation_time
10970
tokens_prompt
77
tokens_completion
406
native_tokens_prompt
93
native_tokens_completion
584
native_tokens_completion_images
(null)
native_tokens_reasoning
62
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
"end_turn"
service_tier
"default"
usage
0.015065
router
(null)
provider_responses
0
endpoint_id
"b0d3beaa-5ff7-4666-a749-b1bb2b14c916"
id
"msg_01RdLm1oqLU5PSF6a2Qm7HFb"
is_byok
false
latency
1098
model_permaslug
"anthropic/claude-4.8-opus-20260528"
provider_name
"Anthropic"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.16.0; linux; x64))"
http_referer
(null)
request_id
"req-1780227656-UBIjyxMS63yYmeZ0DSuV"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1780227656-tV67pchawcH3fF8TpXZf"
upstream_id
"msg_01RdLm1oqLU5PSF6a2Qm7HFb"
total_cost
0.015065
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Anthropic"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
100.00%
Matches word count
n/a
neededClean
false
words
200
90.71%
Dialogue to Total Word Ratio
Ratio: 4.41%, Deviation: 5.59%
neededClean
false
wordsTotal
204
wordsDialogue
9
95.3542%