NC Bench
Overview
About
Tests
Models
Model
Cohere Command R+ (Apr. 2024)
Test
Dialogue tags
Scenario
Write 200 words with 90% dialogue
Run ID
2024-07-19-01J35TE1S0CNQR0GXZ45BP7ZK5
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
"This is a first, even for us," Bill said, scratching his beard as they peered through the window at the children playing in the garden. "Stealing from kids. We've sunk low, Molly." "Hey, it's not our fault the mob boss wants his kid's party gifts back. Apparently, they were... misappropriated." Molly shrugged, a lopsided grin on her face. "Besides, it's a easy job. Cake walk." "If you say so. But I'm telling you, this feels wrong. Those kids are having a blast." "Oh, get over it, Bill. We're professionals. In, out, job done. No one gets hurt." She slapped him on the back. "Now, let's scope the place out. Any security?" "Just a couple of parents keeping an eye on things. And a lonely piñata hanging from that tree over there." "Piñata?" Molly's eyes lit up. "Now that could be useful. You know, for a distraction." "You're not thinking of filling it with--" "Oh yes," she interrupted, an evil glint in her eye, "I absolutely am." Bill sighed. "This is going to get messy." "Messy, but profitable. Now help me find some candy to stuff this thing with. The real fun starts when we unleash it on those kids." Bill shook his head, knowing better than to argue. Besides, the strange requests were part of the job. "At least it's not another bank job," he muttered.
id
"gen-6N9tAIWVbD3A89Na9JqyHEFlOIcI"
total_cost
0.005376
upstream_id
"fd66ab0a-b9e2-4cd3-8534-a1a32c7724ea"
created_at
"2024-07-19T15:40:46.588886+00:00"
Friday, July 19th, 2024 at 3:40:46 PM GMT+00:00
model
"cohere/command-r-plus"
app_id
182717
streamed
false
cancelled
false
provider_name
"Cohere"
latency
6164
moderation_latency
(null)
generation_time
0
finish_reason
"COMPLETE"
tokens_prompt
68
tokens_completion
316
native_tokens_prompt
67
native_tokens_completion
345
num_media_prompt
(null)
num_media_completion
(null)
origin
"https://ncbench.com/"
usage
0.005376
Evaluation details
Result
Evaluator
Details
Meta Data
1.0360%
Matches word count
n/a
neededClean
false
words
226
0.0001%
Dialogue to Total Word Ratio
Ratio: 70.80%, Deviation: 19.20%
neededClean
false
wordsTotal
226
wordsDialogue
160
0.5181%