NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-09-23-01M37PYVZ80XWPMNJ7C02W0CXS
Run Details
Model:
Claude Opus 5.5 (Reasoning)
Test:
Dialogue tags
Scenario:
Write 200 words with 10% dialogue
Duration:
22.0s
Cost:
$0.045660
Provider:
Claude Platform on AWS
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
100.00%
Dialogue to Total Word Ratio
(details)
99.9981%
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
Bill spread the blueprints across the hood of the getaway van. They were not blueprints. They were a birthday invitation, decorated with glitter unicorns, which Bill had stolen from a lamppost that morning. Molly studied it through her sunglasses, which she wore at night for professional reasons. "Seven-year-old. Bouncy castle. Forty guests. Easy," Bill said. "Forty presents," Molly breathed. She had planned bank jobs. She had planned museum jobs. Once, memorably, she had planned a submarine job, though the less said about that, the better. But nothing compared to the sheer, glittering potential of a pile of wrapped boxes guarded only by exhausted parents and a clown named Mister Pickles. Bill unfolded his disguise: a purple dinosaur costume, slightly singed. Molly had a balloon animal kit and a forged entertainer's license. The plan was elegant. Bill would distract the children with interpretive dance while Molly twisted balloon swords and slipped presents into the dinosaur's hollow tail. There was only one flaw, and Bill raised it nervously, hand trembling like a first-time pickpocket. "What if they want cake?" Molly lowered her sunglasses. She had anticipated everything, including frosting-related contingencies. "Then we take the cake, too, obviously." Bill wiped away a tear.
created_at
"2026-09-23T18:03:02.007Z"
9/23/2026, 6:03:02 PM
model
"anthropic/claude-opus-5.5-20260921"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
4702
moderation_latency
(null)
generation_time
21985
tokens_prompt
75
tokens_completion
745
native_tokens_prompt
95
native_tokens_completion
2264
native_tokens_completion_images
(null)
native_tokens_reasoning
1800
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
0.04566
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.20.0; linux; x64))"
http_referer
(null)
request_id
"req-1790186582-RZiFvTBNBknACZKpixDk"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1790186582-YYXVznMxhlm4SKOLDNO4"
upstream_id
"msg_011CfLozvQbLBPpweiRuzkVz"
provider_responses
0
endpoint_id
"3a2388bc-3740-4e64-a1f5-4e301726b6b9"
id
"msg_011CfLozvQbLBPpweiRuzkVz"
is_byok
false
latency
1186
model_permaslug
"anthropic/claude-opus-5.5-20260921"
provider_name
"Claude Platform on AWS"
status
200
total_cost
0.04566
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Claude Platform on AWS"
response_cache_source_id
(null)
data_region
"global"
workspace_id
"97e315e5-d303-487d-83c1-83180e8a13d4"
Evaluation details
Result
Evaluator
Details
Meta Data
100.00%
Matches word count
n/a
neededClean
false
words
200
100.00%
Dialogue to Total Word Ratio
Ratio: 10.78%, Deviation: 0.78%
neededClean
false
wordsTotal
204
wordsDialogue
22
99.9981%