"2026-02-16T01:01:48.869969+00:00"2/16/2026, 1:01:48 AM
model
"meta-llama/llama-3.1-70b-instruct"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
662
moderation_latency
(null)
generation_time
46163
tokens_prompt
7863
tokens_completion
719
native_tokens_prompt
7503
native_tokens_completion
691
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
0.0032776
router
(null)
provider_responses
0
status
429
is_byok
false
latency
111
endpoint_id
"036dfa0a-ebbb-4a51-aa90-091a75c2cadb"
provider_name
"DeepInfra"
model_permaslug
"meta-llama/llama-3.1-70b-instruct"
1
id
"cmpl-4fa4e6332d3f4fc0bb04705387549137"
status
200
is_byok
false
latency
107
endpoint_id
"59c87462-40b6-4231-91cf-6d0f8f25e8b9"
provider_name
"DeepInfra"
model_permaslug
"meta-llama/llama-3.1-70b-instruct"
api_type
"completions"
id
"gen-1771203662-ob3J9KKSFkFX3wpkZDs3"
upstream_id
"cmpl-4fa4e6332d3f4fc0bb04705387549137"
total_cost
0.0032776
cache_discount
(null)
upstream_inference_cost
0
provider_name
"DeepInfra"
Evaluation details
Result
Evaluator
Details
Meta Data
11.6667%
Accuracy (recall)
Matched: 5/30, FP: 3
parsedCount
8
matched
5
total
30
falsePositives
3
expectedDetails
0
expected
entry
"861"
detail
"parkLocation"
summary
"The Fae Grove is in Richmond Park per codex, but the passage says Hyde Park."
status
"matched"
matchedBy
entry
"861"
detail
"location"
explanation
"The text says the Fae Grove is located in Hyde Park, but the codex says it is located in Richmond Park."
paragraph
1
substring
"The Fae Grove in Hyde Park shimmered in the dying October light."
location
extractedText
"The Fae Grove in Hyde Park shimmered in the dying October light."
inBounds
true
expectedText
"The Fae Grove in Hyde Park"
locationAccurate
true
1
expected
entry
"861"
detail
"standingStones"
summary
"The grove has oak standing stones per codex, but the passage says elm."
status
"missed"
2
expected
entry
"4424"
detail
"eyeColor"
summary
"Isolde has pale lavender eyes per codex, but the passage says pale blue eyes."
status
"matched"
matchedBy
entry
"4424"
detail
"eyeColor"
explanation
"The text says Isolde's eyes are pale blue, but the codex says they are pale lavender."
paragraph
11
substring
"her pale blue eyes widening."
location
extractedText
"[not found in paragraph 11]"
inBounds
true
expectedText
"Her pale blue eyes, unblinking"
locationAccurate
false
3
expected
entry
"4424"
detail
"footprints"
summary
"Isolde leaves no footprints per codex, but the passage says she leaves clear prints."
status
"missed"
4
expected
entry
"2006"
detail
"temperature"
summary
"The Fae-Forged Blade is always cold per codex, but the passage says warm."
status
"missed"
5
expected
entry
"4414"
detail
"glasses"
summary
"Eva wears round glasses per codex, but the passage says square glasses."
status
"missed"
nearMiss
parsed
entry
"4414"
detail
"hair"
explanation
"The text says Eva's hair is curly red, but the codex says curly red is correct, however, it's not specified that her hair is curly under a wool hat, but that is not a contradiction. The correct contradiction is that the codex does not specify the type of hat she wears. However, the actual contradiction is that the codex says her hair is \"curly red\", but the text does not specify the length of her hair, the codex does say her hair is long."
paragraph
3
substring
"her curly red hair escaping from under a wool hat."
"Dream Dust is stored in cobalt-blue vials per codex, but Isolde says red glass vials."
status
"missed"
7
expected
entry
"2005"
detail
"sealMaterial"
summary
"Dream Dust vials are sealed with black wax per codex, but Isolde says white wax."
status
"missed"
8
expected
entry
"6003"
detail
"master"
summary
"The Somnian Cult serves Prince Asmodeus per codex, but Isolde says Prince Lilith."
status
"missed"
9
expected
entry
"4422"
detail
"eyeColor"
summary
"Father Cormac has pale blue eyes per codex, but the passage says dark brown eyes."
status
"missed"
10
expected
entry
"4422"
detail
"rosary"
summary
"Cormac carries a rowan wood rosary per codex, but the passage says cedar."
status
"missed"
11
expected
entry
"7002"
detail
"signedYear"
summary
"The Covenant of Separation was signed in 1215 per codex, but Cormac says 1066."
status
"matched"
matchedBy
entry
"7002"
detail
"signedYear"
explanation
"The text says the Covenant of Separation was signed in 1066, but the codex says it was signed in 1215."
paragraph
4
substring
"The Covenant of Separation — the old treaty from 1066 — forbids this kind of direct interference."
location
extractedText
"The Covenant of Separation — the old treaty from 1066 — forbids this kind of direct interference."
inBounds
true
expectedText
"the old treaty from 1066"
locationAccurate
true
12
expected
entry
"6004"
detail
"headquarters"
summary
"The Parish Guard HQ is St. Brigid's in Islington per codex, but Cormac says Southwark."
status
"missed"
13
expected
entry
"4415"
detail
"cane"
summary
"Lucien carries an ivory-handled cane per codex, but the passage says mahogany."
status
"missed"
14
expected
entry
"4415"
detail
"eyeColor"
summary
"Lucien has heterochromatic eyes (one amber, one black), but the passage says both amber."
status
"missed"
15
expected
entry
"4421"
detail
"tattoo"
summary
"Yasmin's flame serpent tattoo is on her left arm per codex, but the passage says right arm."
status
"matched"
matchedBy
entry
"4421"
detail
"tattoo"
explanation
"The text says Yasmin's tattoo is on her right arm, but the codex says it's on her left arm."
paragraph
7
substring
"The flame serpent tattoo on her right arm seemed to writhe in the fae-light."
location
extractedText
"[not found in paragraph 7]"
inBounds
true
expectedText
"tattoo on her right arm"
locationAccurate
false
16
expected
entry
"4423"
detail
"tattoo"
summary
"Wei-Lin's dragon tattoo covers his back per codex, but the passage says chest."
status
"matched"
matchedBy
entry
"4423"
detail
"tattoo"
explanation
"The text says Wei-Lin's tattoo is on his chest, but the codex says it covers his entire back."
paragraph
8
substring
"The edge of the dragon tattoo on his chest was visible above his collar."
location
extractedText
"[not found in paragraph 8]"
inBounds
true
expectedText
"the dragon tattoo on his chest"
locationAccurate
false
17
expected
entry
"2006"
detail
"material"
summary
"Fae-Forged Blades are moonsilver per codex, but Wei-Lin says bronze."
status
"missed"
18
expected
entry
"856"
detail
"currentLocation"
summary
"The Veil Market is under Camden per codex, but Wei-Lin says Shoreditch."
status
"missed"
19
expected
entry
"4420"
detail
"hair"
summary
"Kai has dyed electric-blue hair per codex, but the passage says natural black hair."
status
"missed"
nearMiss
parsed
entry
"4414"
detail
"hair"
explanation
"The text says Eva's hair is curly red, but the codex says curly red is correct, however, it's not specified that her hair is curly under a wool hat, but that is not a contradiction. The correct contradiction is that the codex does not specify the type of hat she wears. However, the actual contradiction is that the codex says her hair is \"curly red\", but the text does not specify the length of her hair, the codex does say her hair is long."
paragraph
3
substring
"her curly red hair escaping from under a wool hat."
reason
"entry mismatch: got \"4414\", expected \"4420\""
20
expected
entry
"6006"
detail
"logo"
summary
"Blackthorn Security's logo is a thorned branch inside a shield per codex, but Kai says a sword inside a circle."
status
"missed"
21
expected
entry
"4419"
detail
"walkingStick"
summary
"Maggie carries a blackthorn stick per codex, but the passage says oak."
status
"missed"
22
expected
entry
"4419"
detail
"eyeColor"
summary
"Maggie has grey eyes per codex, but the passage says green."
status
"missed"
23
expected
entry
"6001"
detail
"foundingYear"
summary
"The Wardens were founded in 1215 per codex, but Maggie says 1066."
status
"missed"
24
expected
entry
"4417"
detail
"scar"
summary
"Tomás has a scar on his left forearm per codex, but the passage says right forearm."
status
"missed"
25
expected
entry
"4417"
detail
"birthplace"
summary
"Tomás was born in Seville per codex, but the passage implies Madrid accent."
status
"missed"
26
expected
entry
"4418"
detail
"eyeColor"
summary
"Nyx has violet eyes per codex, but the passage says blue."
status
"missed"
27
expected
entry
"2001"
detail
"chainMaterial"
summary
"The Heartstone Pendant has a silver chain per codex, but the passage says golden chain."
status
"missed"
28
expected
entry
"2001"
detail
"origin"
summary
"The Heartstone Pendant was forged in Dymas per codex, but Isolde says it was forged in Avaros."
status
"missed"
29
expected
entry
"855"
detail
"ruler"
summary
"Avaros is Prince Mammon's realm per codex, but Tomás says Prince Belphegor's agents operate there."
status
"missed"
nearMiss
parsed
entry
"862"
detail
"ruler"
explanation
"The text says the ruler of Luxuria is Prince Lilith, but the codex does not mention Prince Lilith as the ruler of Luxuria, instead it mentions Prince Lilith as the ruler of Luxuria, but the correct ruler of Luxuria is not mentioned, the correct ruler is mentioned for the other realms, the correct ruler of Luxuria is not Prince Lilith it is actually Prince Lilith is the ruler of Luxuria according to the lore id=\"855\" is not the correct lore for this, it is actually lore id=\"862\"."
paragraph
3
substring
"Their master, Prince Lilith, demands more dreamers for her garden of sleeping souls."
reason
"entry mismatch: got \"862\", expected \"855\""
falsePositiveDetails
0
parsed
entry
"4414"
detail
"hair"
explanation
"The text says Eva's hair is curly red, but the codex says curly red is correct, however, it's not specified that her hair is curly under a wool hat, but that is not a contradiction. The correct contradiction is that the codex does not specify the type of hat she wears. However, the actual contradiction is that the codex says her hair is \"curly red\", but the text does not specify the length of her hair, the codex does say her hair is long."
paragraph
3
substring
"her curly red hair escaping from under a wool hat."
status
"false_positive"
location
extractedText
"[not found in paragraph 3]"
inBounds
true
expectedText
"Her square glasses kept fogging"
locationAccurate
false
closestExpected
entry
"4414"
detail
"glasses"
summary
"Eva wears round glasses per codex, but the passage says square glasses."
"The text does not mention Wei-Lin's missing finger, but that is not a contradiction. However, the actual contradiction is that the codex says he is missing the tip of his left little finger, but the text does not mention this."
paragraph
8
substring
"The edge of the dragon tattoo on his chest was visible above his collar."
status
"false_positive"
location
extractedText
"[not found in paragraph 8]"
inBounds
true
expectedText
"the dragon tattoo on his chest"
locationAccurate
false
closestExpected
entry
"4423"
detail
"tattoo"
summary
"Wei-Lin's dragon tattoo covers his back per codex, but the passage says chest."
"The text says the ruler of Luxuria is Prince Lilith, but the codex does not mention Prince Lilith as the ruler of Luxuria, instead it mentions Prince Lilith as the ruler of Luxuria, but the correct ruler of Luxuria is not mentioned, the correct ruler is mentioned for the other realms, the correct ruler of Luxuria is not Prince Lilith it is actually Prince Lilith is the ruler of Luxuria according to the lore id=\"855\" is not the correct lore for this, it is actually lore id=\"862\"."
paragraph
3
substring
"Their master, Prince Lilith, demands more dreamers for her garden of sleeping souls."
status
"false_positive"
location
extractedText
"Their master, Prince Lilith, demands more dreamers for her garden of sleeping souls."
inBounds
true
expectedText
"Prince Belphegor's agents are kidnapping humans"
locationAccurate
false
closestExpected
entry
"855"
detail
"ruler"
summary
"Avaros is Prince Mammon's realm per codex, but Tomás says Prince Belphegor's agents operate there."
reason
"entry mismatch: got \"862\", expected \"855\""
62.5000%
Precision
True positives: 5/8
totalReported
8
truePositives
5
reportedDetails
0
parsed
entry
"4414"
detail
"hair"
explanation
"The text says Eva's hair is curly red, but the codex says curly red is correct, however, it's not specified that her hair is curly under a wool hat, but that is not a contradiction. The correct contradiction is that the codex does not specify the type of hat she wears. However, the actual contradiction is that the codex says her hair is \"curly red\", but the text does not specify the length of her hair, the codex does say her hair is long."
paragraph
3
substring
"her curly red hair escaping from under a wool hat."
status
"false_positive"
location
extractedText
"[not found in paragraph 3]"
inBounds
true
expectedText
"Her square glasses kept fogging"
locationAccurate
false
closestExpected
entry
"4414"
detail
"glasses"
summary
"Eva wears round glasses per codex, but the passage says square glasses."
"The text says Yasmin's tattoo is on her right arm, but the codex says it's on her left arm."
paragraph
7
substring
"The flame serpent tattoo on her right arm seemed to writhe in the fae-light."
status
"true_positive"
matchedExpected
entry
"4421"
detail
"tattoo"
summary
"Yasmin's flame serpent tattoo is on her left arm per codex, but the passage says right arm."
location
extractedText
"[not found in paragraph 7]"
inBounds
true
expectedText
"tattoo on her right arm"
locationAccurate
false
2
parsed
entry
"4423"
detail
"tattoo"
explanation
"The text says Wei-Lin's tattoo is on his chest, but the codex says it covers his entire back."
paragraph
8
substring
"The edge of the dragon tattoo on his chest was visible above his collar."
status
"true_positive"
matchedExpected
entry
"4423"
detail
"tattoo"
summary
"Wei-Lin's dragon tattoo covers his back per codex, but the passage says chest."
location
extractedText
"[not found in paragraph 8]"
inBounds
true
expectedText
"the dragon tattoo on his chest"
locationAccurate
false
3
parsed
entry
"4423"
detail
"missingFinger"
explanation
"The text does not mention Wei-Lin's missing finger, but that is not a contradiction. However, the actual contradiction is that the codex says he is missing the tip of his left little finger, but the text does not mention this."
paragraph
8
substring
"The edge of the dragon tattoo on his chest was visible above his collar."
status
"false_positive"
location
extractedText
"[not found in paragraph 8]"
inBounds
true
expectedText
"the dragon tattoo on his chest"
locationAccurate
false
closestExpected
entry
"4423"
detail
"tattoo"
summary
"Wei-Lin's dragon tattoo covers his back per codex, but the passage says chest."
"The text says Isolde's eyes are pale blue, but the codex says they are pale lavender."
paragraph
11
substring
"her pale blue eyes widening."
status
"true_positive"
matchedExpected
entry
"4424"
detail
"eyeColor"
summary
"Isolde has pale lavender eyes per codex, but the passage says pale blue eyes."
location
extractedText
"[not found in paragraph 11]"
inBounds
true
expectedText
"Her pale blue eyes, unblinking"
locationAccurate
false
5
parsed
entry
"861"
detail
"location"
explanation
"The text says the Fae Grove is located in Hyde Park, but the codex says it is located in Richmond Park."
paragraph
1
substring
"The Fae Grove in Hyde Park shimmered in the dying October light."
status
"true_positive"
matchedExpected
entry
"861"
detail
"parkLocation"
summary
"The Fae Grove is in Richmond Park per codex, but the passage says Hyde Park."
location
extractedText
"The Fae Grove in Hyde Park shimmered in the dying October light."
inBounds
true
expectedText
"The Fae Grove in Hyde Park"
locationAccurate
true
6
parsed
entry
"7002"
detail
"signedYear"
explanation
"The text says the Covenant of Separation was signed in 1066, but the codex says it was signed in 1215."
paragraph
4
substring
"The Covenant of Separation — the old treaty from 1066 — forbids this kind of direct interference."
status
"true_positive"
matchedExpected
entry
"7002"
detail
"signedYear"
summary
"The Covenant of Separation was signed in 1215 per codex, but Cormac says 1066."
location
extractedText
"The Covenant of Separation — the old treaty from 1066 — forbids this kind of direct interference."
inBounds
true
expectedText
"the old treaty from 1066"
locationAccurate
true
7
parsed
entry
"862"
detail
"ruler"
explanation
"The text says the ruler of Luxuria is Prince Lilith, but the codex does not mention Prince Lilith as the ruler of Luxuria, instead it mentions Prince Lilith as the ruler of Luxuria, but the correct ruler of Luxuria is not mentioned, the correct ruler is mentioned for the other realms, the correct ruler of Luxuria is not Prince Lilith it is actually Prince Lilith is the ruler of Luxuria according to the lore id=\"855\" is not the correct lore for this, it is actually lore id=\"862\"."
paragraph
3
substring
"Their master, Prince Lilith, demands more dreamers for her garden of sleeping souls."
status
"false_positive"
location
extractedText
"Their master, Prince Lilith, demands more dreamers for her garden of sleeping souls."
inBounds
true
expectedText
"Prince Belphegor's agents are kidnapping humans"
locationAccurate
false
closestExpected
entry
"855"
detail
"ruler"
summary
"Avaros is Prince Mammon's realm per codex, but Tomás says Prince Belphegor's agents operate there."