NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-05-29-01KSSK1NBQK9J53Z8RHRF12S0V
Run Details
Model:
Claude Opus 4.8 (Reasoning)
Test:
Dialogue tags
Scenario:
Write 200 words with 90% dialogue
Duration:
9.6s
Cost:
$0.012640
Provider:
Anthropic
Evaluation summary
Score
Evaluator
Details Link
99.98%
Matches word count
(details)
91.12%
Dialogue to Total Word Ratio
(details)
95.5514%
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 really doing this," Molly said. "Stealing presents from children." "Stealing presents from one child. Focus, Bill." "It feels wrong." "It felt wrong when you ate my sandwich last Tuesday, but you survived." "That was different. That was a sandwich." "And this is a PlayStation, three drones, and a gift card to somewhere expensive. Priorities." "What's the plan?" "You dress as the clown." "Absolutely not." "You dress as the clown," Molly repeated, "and distract the kids while I load the loot into the van." "Why am I always the clown?" "Because you have the face for it." "That's hurtful." "It's strategic. Now, the cake comes out at three. That's our window." "What if they cry?" "Children always cry. It's their natural state." "What if the parents notice?" "The parents will be drunk by two-thirty. Trust me, I scouted the invitation." "You scouted the invitation?" "I'm thorough." "Molly, we're stealing from a seven-year-old." "He's turning eight, actually." "That's worse!" "That's irrelevant. Are you in or out?" Bill sighed heavily. "I'm in. But I'm not honking the nose." "You'll honk the nose." "I won't." "You'll honk the nose, Bill." "...I'll honk the nose."
created_at
"2026-05-29T10:04:00.768Z"
5/29/2026, 10:04:00 AM
model
"anthropic/claude-4.8-opus-20260528"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1481
moderation_latency
(null)
generation_time
9563
tokens_prompt
77
tokens_completion
360
native_tokens_prompt
93
native_tokens_completion
487
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
0.01264
router
(null)
provider_responses
0
endpoint_id
"b0d3beaa-5ff7-4666-a749-b1bb2b14c916"
id
"msg_01Jrp9ZJH2pGciDAn3JhBxBf"
is_byok
false
latency
1481
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-1780049040-ulbs1kL0GFsS8vI8EJwE"
session_id
(null)
api_type
"completions"
id
"gen-1780049040-4ZpC9FmTlYPgFCr8jxtv"
upstream_id
"msg_01Jrp9ZJH2pGciDAn3JhBxBf"
total_cost
0.01264
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Anthropic"
response_cache_source_id
(null)
Evaluation details
Result
Evaluator
Details
Meta Data
99.98%
Matches word count
n/a
neededClean
false
words
198
91.12%
Dialogue to Total Word Ratio
Ratio: 95.52%, Deviation: 5.52%
neededClean
false
wordsTotal
201
wordsDialogue
192
95.5514%