NC Bench
Overview
About
Tests
Models
Vendors
Compare
Run: 2026-02-24-01KJ8VWNMBTBCKCNFB8YEWBFZV
Run Details
Model:
Claude Opus 4.6 (Reasoning)
Test:
Dialogue tags
Scenario:
Write 500 words with 50% dialogue
Duration:
21.4s
Cost:
$0.021185
Provider:
Amazon Bedrock
Evaluation summary
Score
Evaluator
Details Link
100.00%
Matches word count
(details)
32.76%
Dialogue to Total Word Ratio
(details)
66.3793%
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
Molly spread the hand-drawn blueprint across the hood of the van, weighing down one corner with a half-eaten burrito and the other with a walkie-talkie that hadn't worked since 2011. The blueprint was done in crayon. She had stolen it from a child, which she considered a warm-up exercise. "Alright, Bill, listen up. The party starts at two. We've got a forty-five-minute window between the arrival of the bounce house and the arrival of the clown." Bill squinted at the blueprint. "Why is everything drawn in orange?" "Because the kid only had one crayon, Bill. Focus. The presents will be staged here, on a folding table near the back patio. We go in through the side gate, grab the loot, and we're out before anyone notices." "What kind of presents are we talking about?" Bill asked, cracking his knuckles with the seriousness of a man preparing to defuse a bomb. "The kid's turning seven. So we're looking at Legos, action figures, maybe a bike if we're lucky. One year I hit a party and got a karaoke machine. Flipped it for forty bucks." "Forty bucks," Bill repeated, his eyes glistening like she had said forty thousand. Molly tapped the blueprint with authority. "Now, the biggest threat is the mother. Name's Dana. She runs the PTA like a paramilitary operation. If she spots us, she will ask us to help set up chairs, and we will never escape." "Can't we just say no?" "You've never met a PTA mom, have you? They don't accept no. They don't even hear no. The word doesn't exist in their vocabulary. You say no and they hand you a bag of streamers and suddenly you're three hours deep into decorating a paper-mâché unicorn." Bill shuddered. He had once been trapped at a bake sale and still bore the emotional scars. "So what's our cover?" he asked. "We're balloon artists. I bought us matching polo shirts." She pulled two purple shirts from the back seat. Each one read Gary's Balloon Bonanza in peeling iron-on letters. "Who's Gary?" "There is no Gary. Gary is a fabrication. Gary is the ghost we hide behind, Bill." He held up the polo shirt and frowned at a mustard stain on the collar. This was not the criminal empire he had envisioned when he dropped out of night school, but rent was rent. Molly rolled up the blueprint and shoved it into her fanny pack with the confidence of a general preparing for invasion. She checked her watch, adjusted her sunglasses, and pointed at the minivan across the street, where a woman was already unloading trays of cupcakes and barking orders into a phone. "There she is. That's Dana. Look at her, Bill. Organized. Ruthless. Probably has a spreadsheet for the goody bags." "She terrifies me," Bill whispered. "Good. Use that fear. Channel it. Now get in the van and put on your polo. We've got presents to liberate." Bill climbed in and silently prayed the Legos would be worth it.
created_at
"2026-02-24T22:21:36.28Z"
2/24/2026, 10:21:36 PM
model
"anthropic/claude-4.6-opus-20260205"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1338
moderation_latency
284
generation_time
20991
tokens_prompt
77
tokens_completion
822
native_tokens_prompt
92
native_tokens_completion
829
native_tokens_completion_images
(null)
native_tokens_reasoning
75
native_tokens_cached
0
num_media_prompt
(null)
num_input_audio_prompt
(null)
num_media_completion
0
num_search_results
(null)
origin
"https://ncbench.com/"
is_byok
false
finish_reason
"stop"
native_finish_reason
"stop"
usage
0.021185
router
(null)
provider_responses
0
endpoint_id
"a8708fb9-08ea-43bb-a3b1-165c87ef09df"
id
"0246004f-696f-4ce8-9e0b-864fdca94a1a"
is_byok
false
latency
1306
model_permaslug
"anthropic/claude-4.6-opus-20260205"
provider_name
"Amazon Bedrock"
status
200
api_type
"completions"
id
"gen-1771971696-iR8sETvfQDcXrgFLlGgJ"
upstream_id
"0246004f-696f-4ce8-9e0b-864fdca94a1a"
total_cost
0.021185
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Amazon Bedrock"
Evaluation details
Result
Evaluator
Details
Meta Data
100.00%
Matches word count
n/a
neededClean
false
words
500
32.76%
Dialogue to Total Word Ratio
Ratio: 54.72%, Deviation: 4.72%
neededClean
false
wordsTotal
508
wordsDialogue
278
66.3793%