NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-06-01-01KT1JA296YE3QVYGZV0JN9YFM
Run Details
Model:
Claude Opus 4.8 (Reasoning, Low)
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
9.8s
Cost:
$0.012990
Provider:
Anthropic
Evaluation summary
Score
Evaluator
Details Link
86.38%
Matches word count
(details)
67.96%
Dialogue to Total Word Ratio
(details)
77.1693%
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
"So we're really doing this?" Bill asked. "We're doing this," Molly said. "Stealing presents. From children." "From a child. Singular. One spoiled birthday boy." "How do we get in?" "Clown costumes. Nobody questions clowns." "I hate clowns." "Everyone hates clowns, Bill. That's the genius. They won't look at us." "What about the parents?" "Distracted. Cake, juice boxes, existential despair. We slip in, grab the pile, slip out." "And the kids?" "What about them?" "They'll cry." "Children cry constantly. It's their natural state." "This feels wrong, Molly." "You robbed a retirement home last spring." "That was different. They were asleep." "You stole from sleeping seniors but won't rob a party?" "Seniors don't believe in magic anymore. Kids do." "Oh, here we go." "I'm just saying, there's an innocence—" "There's an Xbox, Bill. Three of them, probably. Plus cash in birthday cards." "How much cash?" "Now you're interested." "I'm asking questions." "Grandparents give twenties. There'll be a dozen cards minimum." "So roughly—" "Two hundred, easy. Plus electronics." "Fine. But I'm not making balloon animals." "You'll make balloon animals if a child asks." "I don't know how!" "Then learn, Bill. We're professionals."
created_at
"2026-06-01T12:25:03.022Z"
6/1/2026, 12:25:03 PM
model
"anthropic/claude-4.8-opus-20260528"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1479
moderation_latency
(null)
generation_time
9744
tokens_prompt
77
tokens_completion
363
native_tokens_prompt
93
native_tokens_completion
501
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
"end_turn"
service_tier
"default"
usage
0.01299
router
(null)
provider_responses
0
endpoint_id
"b0d3beaa-5ff7-4666-a749-b1bb2b14c916"
id
"msg_01B9dKYCwSj135tLZjqPBkWW"
is_byok
false
latency
1479
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-1780316703-Wol2xYPtIeqSrmoplqGv"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1780316703-cCMlrWd7PEJ0MlSaQQA5"
upstream_id
"msg_01B9dKYCwSj135tLZjqPBkWW"
total_cost
0.01299
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Anthropic"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
86.38%
Matches word count
n/a
neededClean
false
words
189
67.96%
Dialogue to Total Word Ratio
Ratio: 97.88%, Deviation: 7.88%
neededClean
false
wordsTotal
189
wordsDialogue
185
77.1693%