NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-10-08-01M4DA745K1Y5XHPYX92AW8CYV
Run Details
Model:
Claude Haiku 5.5 (Adaptive)
Test:
Dialogue tags
Scenario:
Write unattributed dialogue
Duration:
6.2s
Cost:
$0.000500
Provider:
Claude Platform on AWS
Evaluation summary
Score
Evaluator
Details Link
13.53%
Count dialogue tags
(details)
Model input
User/Human
You are an expert novel writer in the cozy mystery genre. Write a scene of about 200 words, with dialogue between two characters. The dialogue should be unattributed, meaning the speaker is not explicitly mentioned. The scene should be set in a small town bakery. Only return the story, nothing else.
Model response
Messages
Meta
AI
The bell above the door jingled, and the cinnamon rolls on the back shelf seemed to hold their breath. "You're late,"
Marigold said
, wiping flour from her hands. "Third Tuesday in a row." "Traffic on Maple. Someone parked a tractor across the whole street." "Mm. Or someone didn't want to be seen walking into a bakery at six in the morning." "Don't start." "I'm not starting anything. I'm just noticing that the cash drawer is empty and you've got a smear of blue icing on your collar." "Blue icing is for the Harwick christening cake. You know that." "I know the Harwick cake was ordered for Saturday. Yet here it is, Thursday, and the lid's already off the tin." A long pause. The oven timer ticked down. "Keep your voice low. Did you bring the envelope?" "Did you bring the money?" "Half of it. The rest after I see what's inside." "You don't trust me." "I trust you about as far as I can throw that rolling pin. Which, after the last month, is a fair distance." A soft laugh, then a rustle of paper sliding across the marble counter. "Careful. Those are the only copies." "Copies of what?" "Of the recipe book. The one Edith's family has been hiding since 1952." Silence. Then the sharp click of a lock turning in the back room. "Well,"
Marigold said
quietly, "now somebody knows we have it."
created_at
"2026-10-08T08:31:29.467Z"
10/8/2026, 8:31:29 AM
model
"anthropic/claude-haiku-5.5-20261007"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
6222
moderation_latency
(null)
generation_time
6222
tokens_prompt
78
tokens_completion
339
native_tokens_prompt
109
native_tokens_completion
979
native_tokens_completion_images
(null)
native_tokens_reasoning
495
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
0.0005004
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.21.0; linux; x64))"
http_referer
(null)
request_id
"req-1791448289-Pzwv1siei90uVYrF7N3b"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1791448289-3VS4SwtfYZ5Wu9ghOhap"
upstream_id
"msg_011CfpTYokYZcjXjCN7ka3xJ"
provider_responses
0
endpoint_id
"b6cdf493-3a61-441f-a134-e7944fddb980"
id
"msg_011CfpTYokYZcjXjCN7ka3xJ"
is_byok
false
latency
782
model_permaslug
"anthropic/claude-haiku-5.5-20261007"
provider_name
"Claude Platform on AWS"
status
200
total_cost
0.0005004
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
13.53%
Count dialogue tags
n/a
neededClean
false
dialogueTags
0
"Marigold said"
1
"Marigold said quietly"