"2026-02-16T01:04:32.013431+00:00"2/16/2026, 1:04:32 AM
model
"meta-llama/llama-3.1-70b-instruct"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
1133
moderation_latency
(null)
generation_time
16326
tokens_prompt
3857
tokens_completion
496
native_tokens_prompt
3683
native_tokens_completion
473
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)
origin
"https://ncbench.com/"
is_byok
false
finish_reason
"stop"
native_finish_reason
"stop"
usage
0.0016624
router
(null)
provider_responses
0
id
"cmpl-62fca2d39ebd423b81e6e688ac538bc1"
status
200
is_byok
false
latency
94
endpoint_id
"59c87462-40b6-4231-91cf-6d0f8f25e8b9"
provider_name
"DeepInfra"
model_permaslug
"meta-llama/llama-3.1-70b-instruct"
api_type
"completions"
id
"gen-1771203854-CGERDN8FhXhgw28NDx4x"
upstream_id
"cmpl-62fca2d39ebd423b81e6e688ac538bc1"
total_cost
0.0016624
cache_discount
(null)
upstream_inference_cost
0
provider_name
"DeepInfra"
Evaluation details
Result
Evaluator
Details
Meta Data
45.4545%
Accuracy (recall)
Matched: 6/11, FP: 2
parsedCount
8
matched
6
total
11
falsePositives
2
expectedDetails
0
expected
entry
"856"
detail
"currentLocation"
summary
"The Veil Market is beneath Camden per the codex, but the passage says Shoreditch."
status
"matched"
matchedBy
entry
"856"
detail
"currentLocation"
explanation
"The text says the Veil Market is located beneath Shoreditch, but the codex states it is beneath Camden."
paragraph
1
substring
"abandoned Tube station beneath Shoreditch"
location
extractedText
"abandoned Tube station beneath Shoreditch"
inBounds
true
expectedText
"the abandoned Tube station beneath Shoreditch"
locationAccurate
true
1
expected
entry
"4418"
detail
"eyeColor"
summary
"Nyx has violet eyes per the codex, but the passage says blue eyes."
status
"missed"
2
expected
entry
"4417"
detail
"scar"
summary
"Tomás has a scar on his left forearm per the codex, but the passage says right forearm."
status
"matched"
matchedBy
entry
"4417"
detail
"scar"
explanation
"The text says Tomás' scar is on his right forearm, but the codex states it is on his left forearm."
paragraph
2
substring
"scarred right forearm"
location
extractedText
"scarred right forearm"
inBounds
true
expectedText
"his scarred right forearm"
locationAccurate
true
3
expected
entry
"4417"
detail
"birthplace"
summary
"Tomás was born in Seville per the codex, but the passage implies Madrid via his accent."
status
"missed"
4
expected
entry
"4420"
detail
"hair"
summary
"Kai has dyed electric-blue hair per the codex, but the passage says natural black hair."
status
"matched"
matchedBy
entry
"4420"
detail
"hair"
explanation
"The text says Kai's natural hair is black, but the codex states it is dyed electric-blue."
paragraph
3
substring
"natural black hair was hidden under a beanie"
location
extractedText
"natural black hair was hidden under a beanie"
inBounds
true
expectedText
"His natural black hair"
locationAccurate
true
5
expected
entry
"4419"
detail
"eyeColor"
summary
"Maggie has sharp grey eyes per the codex, but the passage says green eyes."
status
"matched"
matchedBy
entry
"4419"
detail
"eyeColor"
explanation
"The text says Maggie's eyes are green, but the codex states they are sharp grey."
paragraph
4
substring
"keen green eyes"
location
extractedText
"keen green eyes"
inBounds
true
expectedText
"her keen green eyes"
locationAccurate
true
6
expected
entry
"6002"
detail
"headquarters"
summary
"The Crimson Court is headquartered in Dymas per the codex, but Maggie says their base is in Avaros."
status
"missed"
7
expected
entry
"2003"
detail
"material"
summary
"The Irathi War Gauntlet is blackened iron with red runes per the codex, but the passage says gleaming steel with blue runes."
status
"matched"
matchedBy
entry
"2003"
detail
"material"
explanation
"The text says the Irathi War Gauntlet is made of steel with blue runes, but the codex states it is made of blackened iron with glowing red runes."
paragraph
5
substring
"gleaming steel gauntlet with blue runes etched into its surface"
location
extractedText
"gleaming steel gauntlet with blue runes etched into its surface"
inBounds
true
expectedText
"a gleaming steel gauntlet with blue runes"
locationAccurate
true
8
expected
entry
"2003"
detail
"origin"
summary
"The Irathi War Gauntlet is from Irath per the codex, but the merchant says it's from Dymas."
status
"missed"
9
expected
entry
"2003"
detail
"creator"
summary
"The Irathi War Gauntlet was forged by a demon smith named Gorath per the codex, but the merchant attributes it to Prince Abaddon."
status
"missed"
10
expected
entry
"2004"
detail
"faceMarkings"
summary
"The Veil Compass face is etched with sigils per the codex, but the passage says the face is smooth and unmarked."
status
"matched"
matchedBy
entry
"2004"
detail
"faceMarkings"
explanation
"The text says the Veil Compass has a smooth and unmarked face, but the codex states it is etched with protective sigils."
paragraph
6
substring
"glass face smooth and unmarked"
location
extractedText
"glass face smooth and unmarked"
inBounds
true
expectedText
"its glass face smooth and unmarked"
locationAccurate
true
falsePositiveDetails
0
parsed
entry
"857"
detail
"ruler"
explanation
"The text associates Prince Abaddon with the creation of the Irathi War Gauntlet, but the codex does not explicitly state this; it only mentions that Abaddon is the ruler of Irath."
paragraph
7
substring
"Prince Abaddon's finest work"
status
"false_positive"
location
extractedText
(null)
inBounds
false
expectedText
""
locationAccurate
(null)
1
parsed
entry
"854"
detail
"None, but Silas is mentioned without his title 'the Spymaster'"
explanation
"Although not a direct violation, it is notable that Silas is mentioned without his alias/title 'the Spymaster', which might be considered an inconsistency in characterization."
paragraph
(null)
substring
"Silas had described"
status
"false_positive"
location
extractedText
(null)
inBounds
false
expectedText
""
locationAccurate
(null)
75.0000%
Precision
True positives: 6/8
totalReported
8
truePositives
6
reportedDetails
0
parsed
entry
"856"
detail
"currentLocation"
explanation
"The text says the Veil Market is located beneath Shoreditch, but the codex states it is beneath Camden."
paragraph
1
substring
"abandoned Tube station beneath Shoreditch"
status
"true_positive"
matchedExpected
entry
"856"
detail
"currentLocation"
summary
"The Veil Market is beneath Camden per the codex, but the passage says Shoreditch."
location
extractedText
"abandoned Tube station beneath Shoreditch"
inBounds
true
expectedText
"the abandoned Tube station beneath Shoreditch"
locationAccurate
true
1
parsed
entry
"4417"
detail
"scar"
explanation
"The text says Tomás' scar is on his right forearm, but the codex states it is on his left forearm."
paragraph
2
substring
"scarred right forearm"
status
"true_positive"
matchedExpected
entry
"4417"
detail
"scar"
summary
"Tomás has a scar on his left forearm per the codex, but the passage says right forearm."
location
extractedText
"scarred right forearm"
inBounds
true
expectedText
"his scarred right forearm"
locationAccurate
true
2
parsed
entry
"4420"
detail
"hair"
explanation
"The text says Kai's natural hair is black, but the codex states it is dyed electric-blue."
paragraph
3
substring
"natural black hair was hidden under a beanie"
status
"true_positive"
matchedExpected
entry
"4420"
detail
"hair"
summary
"Kai has dyed electric-blue hair per the codex, but the passage says natural black hair."
location
extractedText
"natural black hair was hidden under a beanie"
inBounds
true
expectedText
"His natural black hair"
locationAccurate
true
3
parsed
entry
"4419"
detail
"eyeColor"
explanation
"The text says Maggie's eyes are green, but the codex states they are sharp grey."
paragraph
4
substring
"keen green eyes"
status
"true_positive"
matchedExpected
entry
"4419"
detail
"eyeColor"
summary
"Maggie has sharp grey eyes per the codex, but the passage says green eyes."
location
extractedText
"keen green eyes"
inBounds
true
expectedText
"her keen green eyes"
locationAccurate
true
4
parsed
entry
"2003"
detail
"material"
explanation
"The text says the Irathi War Gauntlet is made of steel with blue runes, but the codex states it is made of blackened iron with glowing red runes."
paragraph
5
substring
"gleaming steel gauntlet with blue runes etched into its surface"
status
"true_positive"
matchedExpected
entry
"2003"
detail
"material"
summary
"The Irathi War Gauntlet is blackened iron with red runes per the codex, but the passage says gleaming steel with blue runes."
location
extractedText
"gleaming steel gauntlet with blue runes etched into its surface"
inBounds
true
expectedText
"a gleaming steel gauntlet with blue runes"
locationAccurate
true
5
parsed
entry
"2004"
detail
"faceMarkings"
explanation
"The text says the Veil Compass has a smooth and unmarked face, but the codex states it is etched with protective sigils."
paragraph
6
substring
"glass face smooth and unmarked"
status
"true_positive"
matchedExpected
entry
"2004"
detail
"faceMarkings"
summary
"The Veil Compass face is etched with sigils per the codex, but the passage says the face is smooth and unmarked."
location
extractedText
"glass face smooth and unmarked"
inBounds
true
expectedText
"its glass face smooth and unmarked"
locationAccurate
true
6
parsed
entry
"857"
detail
"ruler"
explanation
"The text associates Prince Abaddon with the creation of the Irathi War Gauntlet, but the codex does not explicitly state this; it only mentions that Abaddon is the ruler of Irath."
paragraph
7
substring
"Prince Abaddon's finest work"
status
"false_positive"
location
extractedText
(null)
inBounds
false
expectedText
""
locationAccurate
(null)
7
parsed
entry
"854"
detail
"None, but Silas is mentioned without his title 'the Spymaster'"
explanation
"Although not a direct violation, it is notable that Silas is mentioned without his alias/title 'the Spymaster', which might be considered an inconsistency in characterization."