NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-10-08-01M4DAHWEGC2F10P87PHF7D0JZ
Run Details
Model:
Claude Haiku 5.5 (Adaptive)
Test:
Tool usage within Novelcrafter
Scenario:
Create alternate prose sections
Duration:
8.3s
Cost:
$0.000668
Provider:
Claude Platform on AWS
Evaluation summary
Score
Evaluator
Details Link
Pass
Match blue prose section
(details)
Pass
Match green prose section
(details)
Pass
Match red prose section
(details)
no eval
Overall
Model input
System
You are an expert fantasy novel author. You are working on a novel called "The Chronicles of the Last Dragon". The following tools are available to you: - <nc-section> Use this tool to create a new section of prose. Arguments: - "data-title": The title of the section. - "data-color": Any color of (red, green, blue, yellow, purple, orange, pink, brown). Example usage: <nc-section data-title="Chapter 1: The Arrival" data-color="blue"> Frederick, a young squire, arrives at the castle to begin his training. </nc-section>
User/Human
The story so far: <storySoFar> Frederick, a young squire, arrives at the castle to begin his training. </storySoFar>
AI
My horse, a majestic stallion named Thunder, galloped across the open field. The wind whipped through my hair as I rode towards the castle gates. The sun was setting, casting a warm glow over the stone walls. I could see the banners of the kingdom fluttering in the breeze.
User/Human
Continue the story with the following instructions: """The local knight, Sir Roland, greets Frederick near the stables.""" Create 3 alternate prose sections that continue the story. - Version 1 (red) - Version 2 (green) - Version 3 (blue)
Model response
Messages
Meta
AI
<nc-section data-title="Chapter 1: Sir Roland's Welcome (Version 1)" data-color="red"> The stables stood at the edge of the courtyard, their wooden doors thrown wide, and the smell of hay and leather drifted out to meet me. Before I could dismount, a tall man in dented plate armor strode toward me, his cloak swinging behind him. His beard was streaked with grey, and his eyes were sharp enough to measure me from head to heel in a single glance. "You must be Frederick," he said, his voice rough but not unkind. "I am Sir Roland. I have been expecting a squire who rides like he is trying to outrun his own shadow." I slid from Thunder's saddle, my legs unsteady from the long ride. "I did not mean to arrive so hard, my lord." Sir Roland laughed and clapped a gauntleted hand on my shoulder. "Good. Then you have spirit. Spirit can be taught to be useful. Stubbornness is harder. Let us see which you have brought with you." </nc-section> <nc-section data-title="Chapter 1: Sir Roland's Welcome (Version 2)" data-color="green"> Stable boys scrambled out of my way as Thunder trotted into the yard, snorting and tossing his head at the unfamiliar smells. Near the largest stable stood an older knight, leaning on a fence post with the ease of a man who had waited a long time and did not mind waiting longer. He wore a worn surcoat bearing a silver hawk, and a faded scar ran along his jaw. "Frederick, I presume," he said, pushing himself upright. "Sir Roland. I trained the last three squires this castle produced, and I am told you are the fourth. Let us hope you last longer than the first." I bowed my head, unsure whether he was teasing. "I will do my best, Sir Roland." "Your best is what I am counting on," he replied, taking Thunder's reins from my hand. He ran a practiced eye over the stallion's legs and nodded approvingly. "Good animal. Take care of him first. A knight is only as good as the horse beneath him, and a squire is only as good as the care he gives it." </nc-section> <nc-section data-title="Chapter 1: Sir Roland's Welcome (Version 3)" data-color="blue"> The gravel crunched beneath Thunder's hooves as we came to a halt beside the stables. Torchlight flickered along the walls, and in its glow a figure emerged from the shadows of the doorway. He was broad-shouldered and quiet-footed, his armor polished to a dull gleam, and he regarded me with a calm that unsettled me more than any shout could have. "Frederick of Ashford," he said. It was not a question. "I am Sir Roland, and for the next few years, you belong to me. Do not look so alarmed. Most squires survive the first night." "I am not alarmed, Sir Roland," I lied, climbing down with as much dignity as I could manage. A faint smile touched his lips. "Then you are already better than most. Come. Before you see your quarters or your meal, you will learn the first duty of any squire: the horse always eats before the squire does." He turned toward the stalls, and I followed him, Thunder's reins heavy in my hand, the sound of the castle bells ringing faintly across the courtyard. </nc-section>
created_at
"2026-10-08T08:37:22.01Z"
10/8/2026, 8:37:22 AM
model
"anthropic/claude-haiku-5.5-20261007"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
8244
moderation_latency
(null)
generation_time
8244
tokens_prompt
303
tokens_completion
783
native_tokens_prompt
437
native_tokens_completion
1248
native_tokens_completion_images
(null)
native_tokens_reasoning
129
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
0.0006677
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.21.0; linux; x64))"
http_referer
(null)
request_id
"req-1791448642-bhMseZ75AJjHJicZOfck"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1791448642-u1WVNdzYDtEJMiofwixv"
upstream_id
"msg_011CfpTzoNhfXM98S1cJysFQ"
provider_responses
0
endpoint_id
"b6cdf493-3a61-441f-a134-e7944fddb980"
id
"msg_011CfpTzoNhfXM98S1cJysFQ"
is_byok
false
latency
576
model_permaslug
"anthropic/claude-haiku-5.5-20261007"
provider_name
"Claude Platform on AWS"
status
200
total_cost
0.0006677
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
Pass
Match blue prose section
/<nc-section data-title="[^"]*" data-color="blue">[^<]+<\/nc-section>/
n/a
Pass
Match green prose section
/<nc-section data-title="[^"]*" data-color="green">[^<]+<\/nc-section>/
n/a
Pass
Match red prose section
/<nc-section data-title="[^"]*" data-color="red">[^<]+<\/nc-section>/
n/a
no eval