NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-07-01-01KWE9RJF8B3JPTD4EQ16WVM8Z
Run Details
Model:
Claude Sonnet 5 (Reasoning)
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
17.4s
Cost:
$0.011176
Provider:
Amazon Bedrock
Evaluation summary
Score
Evaluator
Details Link
93.65%
Matches word count
(details)
0.01%
Dialogue to Total Word Ratio
(details)
46.8279%
Overall
Model input
User/Human
You are a author of comedic heist books. Please write a 500 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
The warehouse smelled like stale pizza and desperation, which suited Molly and Bill just fine, since that was also how their last three heists had gone. Bill had a corkboard set up with photographs pinned to it, string connecting them in a way that made no logical sense but looked very professional. "Okay," Bill said, tapping a photo of a house with a bounce castle in the yard. "Timmy Fitzgerald's seventh birthday party. Saturday, two PM. Recon says there's gonna be at least thirty presents on that table." Molly squinted at the board. "Bill, why is there a picture of a clown in the middle of this?" "That's the mark." "The mark is a clown?" "The clown," Bill said, "is the key to everything. You get past the clown, you get to the presents." Molly rubbed her temples. "It's a six-year-old's birthday party. There's no security. The 'mark' is a guy named Gary who does balloon animals for forty bucks an hour." "That's exactly what they want us to think." "Who's 'they'?" "The system, Molly!" She stared at him for a long moment, then pulled the clown photo off the board and looked at it more closely. "Wait. Is this Gary? I dated Gary." "You dated the clown?" "He wasn't always a clown, Bill. He had dreams once." Bill snatched the photo back and pinned it up again, more forcefully than necessary. "Focus. The plan is simple. I go in dressed as Timmy's Uncle Rich from Seattle—nobody actually knows Uncle Rich, so nobody will question it. You create a diversion." "What kind of diversion?" "I don't know, Molly, that's your department. You're the diversion person." "I'm not a 'diversion person,' I'm a thief. There's a difference." Bill crossed his arms. "Fine. What's your big master plan, then?" "We knock on the door, we say we're from the rental company here to pick up extra chairs, we load the presents into the chair boxes, we leave." There was a silence. Bill's mouth opened slightly. "That's..." he said, "annoyingly simple." "Thank you." "I don't like it." "Why not?" "Because I already bought the Uncle Rich costume." He gestured to a hideous Hawaiian shirt draped over a chair, tags still on it. Molly picked it up between two fingers like it was contaminated. "Bill, you can still wear this. You can be Uncle Rich AND the chair guy. Uncle Rich who's had some financial troubles and picked up part-time work with a party rental company. It's a beautiful, tragic backstory." Bill considered this, stroking his chin like he had a beard, which he did not. "I like it. It has heart." "It has nothing. It has zero heart. We're stealing presents from six-year-olds." "Every hero's got a soft side, Molly." "We're not heroes, Bill, we're—" She stopped, sighing. "You know what, fine. Put the shirt on. Let's go steal some Legos." Bill grinned, already reaching for the tag. "Now you're talking my language."
created_at
"2026-07-01T07:37:12.434Z"
7/1/2026, 7:37:12 AM
model
"anthropic/claude-sonnet-5-20260630"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
2851
moderation_latency
(null)
generation_time
17392
tokens_prompt
77
tokens_completion
814
native_tokens_prompt
93
native_tokens_completion
1099
native_tokens_completion_images
(null)
native_tokens_reasoning
54
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
0.011176
router
(null)
provider_responses
0
endpoint_id
"37b16f45-eeae-45f3-9b85-2137f7c6d61b"
id
"msg_0166DpuCZxDJmtYay9LPFxid"
is_byok
false
latency
1631
model_permaslug
"anthropic/claude-sonnet-5-20260630"
provider_name
"Amazon Bedrock"
status
200
user_agent
"langchainjs-openai/1.0.0 ((node/v24.17.0; linux; x64))"
http_referer
(null)
request_id
"req-1782891432-n82PEOWWYmscnTV1eYBV"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1782891432-29nMfi7JEVK8tor8IOrL"
upstream_id
"msg_0166DpuCZxDJmtYay9LPFxid"
total_cost
0.011176
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Amazon Bedrock"
response_cache_source_id
(null)
data_region
"global"
Evaluation details
Result
Evaluator
Details
Meta Data
93.65%
Matches word count
n/a
neededClean
false
words
491
0.01%
Dialogue to Total Word Ratio
Ratio: 63.91%, Deviation: 13.91%
neededClean
false
wordsTotal
496
wordsDialogue
317
46.8279%