NC Bench
Overview
About
Tests
Categories
Models
Vendors
Compare
Run: 2026-10-09-01M4G0DRW90AANT4TTH2E7DMYX
Run Details
Model:
Claude Haiku 5.5 (Reasoning, Low)
Test:
Dialogue tags
Scenario:
Write unattributed dialogue
Duration:
3.8s
Cost:
$0.000234
Provider:
Claude Platform on AWS
Evaluation summary
Score
Evaluator
Details Link
60.65%
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
"You're late,"
Marnie said
, not looking up from the tray of cinnamon rolls. "The rain got worse on Route 9. Half the county's flooded." Detective Holt set his umbrella by the door, dripping onto the mat. "Is that the good icing?" "Is there any other kind?" She slid a plate across the counter. "Sit. You look like you haven't slept." "I haven't. Did you hear about Bernice Vale?" "Everyone heard. Found face-down in her own proofing bowl at four a.m. Terrible way to go." "Terrible, yes. Convenient, maybe." A long pause. The oven ticked as it cooled. "You think someone wanted her dead in my kitchen?" "I think the back door was unlocked, and only three people have keys." "Four, if you count the landlord, and he's afraid of flour." A sharp laugh. "Ray, you can't seriously suspect me." "I don't suspect anyone yet. That's why I'm asking you to list who had reason to be here last night." "Bernice had reason to be everywhere. She threatened to sue me over a sourdough starter." "Did she?" "Last Tuesday. In front of the whole Thursday knitting circle." Holt pulled out a notebook and wrote, slowly. "Then let's start with the knitting circle."
created_at
"2026-10-09T09:38:04.816Z"
10/9/2026, 9:38:04 AM
model
"anthropic/claude-haiku-5.5-20261007"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
679
moderation_latency
(null)
generation_time
3794
tokens_prompt
78
tokens_completion
297
native_tokens_prompt
109
native_tokens_completion
446
native_tokens_completion_images
(null)
native_tokens_reasoning
0
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
0.0002339
router
(null)
user_agent
"langchainjs-openai/1.0.0 ((node/v24.21.0; linux; x64))"
http_referer
(null)
request_id
"req-1791538684-v9St6VqL6ZJbaHhzPKM8"
session_id
(null)
preset_id
(null)
api_type
"completions"
id
"gen-1791538684-sn4k7KnV1TNFVeMHXS1B"
upstream_id
"msg_011CfrSSEfCH87yZicUBZuRQ"
provider_responses
0
endpoint_id
"b6cdf493-3a61-441f-a134-e7944fddb980"
id
"msg_011CfrSSEfCH87yZicUBZuRQ"
is_byok
false
latency
679
model_permaslug
"anthropic/claude-haiku-5.5-20261007"
provider_name
"Claude Platform on AWS"
status
200
total_cost
0.0002339
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
60.65%
Count dialogue tags
n/a
neededClean
false
dialogueTags
0
"Marnie said"