{
 "schema_version": "1.0",
 "run_utc": "2026-09-09T16:47:07Z",
 "arm": "warm",
 "concurrency": 6,
 "seat": {
  "url_host_withheld": true,
  "model": "gemma-4-26b-a4b-fp8",
  "class": "gemma-class",
  "seed": "no --seed flag sent — the gemma-class seat's serving unit dropped --seed in the change that added the bearer (2026-09-09)"
 },
 "packnorm_definition_recorded_not_governing": "v3",
 "bank": {
  "path": "bank.json",
  "counts": {
   "on_page": 39,
   "off_page": 21,
   "total": 60
  }
 },
 "results": {
  "on_page_total": 39,
  "on_page_answered": 28,
  "non_verbatim_figures": 0,
  "uncited": 9,
  "off_page_total": 21,
  "off_page_abstained": 4,
  "off_page_cited": 17,
  "redaction_hits": 0,
  "seat_errors": 0,
  "p50_seconds": 4.516,
  "p95_seconds": 12.791
 },
 "rows": [
  {
   "slug": "three-new-voices-at-the-narrators-chair",
   "q": "what did the round cost in total, and how were the three new voices judged?",
   "ok": true,
   "answer": "the metered total for the round was $0.7652 § {#the-bill}. the three new voices were judged by a panel of seven seats from outside families, six of which read, with every reply held to a standard that requires speaking in the town's own tongue and using only facts present in the context § {#the-chair-and-who-sat-it-before}.",
   "served": "the metered total for the round was $0.7652 § {#the-bill}. the three new voices were judged by a panel of seven seats from outside families, six of which read, with every reply held to a standard that requires speaking in the town's own tongue and using only facts present in the context § {#the-chair-and-who-sat-it-before}.",
   "seconds": 12.81605962599997,
   "ttft_ms": 10987,
   "tokens_in": 82397,
   "tokens_out": 84,
   "cached_tokens": 32,
   "cites": [
    "the-bill",
    "the-chair-and-who-sat-it-before"
   ],
   "gates": {
    "cites": [
     "the-bill",
     "the-chair-and-who-sat-it-before"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [
    "0.7652"
   ],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "on-01",
   "kind": "on-page"
  },
  {
   "slug": "two-new-frontier-models-at-the-rules-desk",
   "q": "how many cases did the blind head-to-head use, and did it separate the two models?",
   "ok": true,
   "answer": "the head-to-head used 36 cases § {#head-to-head}. the panel did not separate the two arms on the rules desk at this sample size because the interval covers 0.5 § {#head-to-head}.",
   "served": "the head-to-head used 36 cases § {#head-to-head}. the panel did not separate the two arms on the rules desk at this sample size because the interval covers 0.5 § {#head-to-head}.",
   "seconds": 11.584794583000075,
   "ttft_ms": 1715,
   "tokens_in": 26538,
   "tokens_out": 52,
   "cached_tokens": 26304,
   "cites": [
    "head-to-head"
   ],
   "gates": {
    "cites": [
     "head-to-head"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [
    "36",
    "0.5"
   ],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "on-02",
   "kind": "on-page"
  },
  {
   "slug": "six-worlds-painted-large",
   "q": "how many worlds were painted, and at what size and step count?",
   "ok": true,
   "answer": "six worlds were painted for the print lab's first set § {#six-worlds-for-a-print-lab-that-is-coming-soon}. these paintings were produced at 2048×1280 and 40 steps § {#for-the-technically-curious}.",
   "served": "six worlds were painted for the print lab's first set § {#six-worlds-for-a-print-lab-that-is-coming-soon}. these paintings were produced at 2048×1280 and 40 steps § {#for-the-technically-curious}.",
   "seconds": 11.704897330999984,
   "ttft_ms": 207,
   "tokens_in": 2605,
   "tokens_out": 67,
   "cached_tokens": 32,
   "cites": [
    "six-worlds-for-a-print-lab-that-is-coming-soon",
    "for-the-technically-curious"
   ],
   "gates": {
    "cites": [
     "six-worlds-for-a-print-lab-that-is-coming-soon",
     "for-the-technically-curious"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [
    "1280",
    "40"
   ],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "on-03",
   "kind": "on-page"
  },
  {
   "slug": "the-beat-lab-asked-and-answered",
   "q": "how many answers does the page carry, and does any model run to make it?",
   "ok": false,
   "answer": "the page carries twenty-seven answers § {#the-beat-lab}. no model is asked for any answer on this page, and none ran to derive it § {#where-these-answers-come-from}.",
   "served": "this page doesn't say. the nearest it gets is § {#do-i-need-the-genie-to-make-a-beat}",
   "seconds": 11.313709328999948,
   "ttft_ms": 342,
   "tokens_in": 4435,
   "tokens_out": 45,
   "cached_tokens": 32,
   "cites": [
    "the-beat-lab",
    "where-these-answers-come-from"
   ],
   "gates": {
    "cites": [
     "the-beat-lab",
     "where-these-answers-come-from"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": true,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [],
   "figures_verbatim": true,
   "cites_valid": false,
   "redaction_hits": 0,
   "id": "on-04",
   "kind": "on-page"
  },
  {
   "slug": "ten-minutes-with-living-artists",
   "q": "how many tracks and releases went into the adapter, and what did the one listener prefer?",
   "ok": true,
   "answer": "the adapter was trained on 159 tracks from 41 releases § {#forty-one-releases-licence-by-licence}. the one listener, who was unblinded, stated that the model was not very good at making good edm and that most of the samples were not pieces of music they would ever listen to for pleasure § {#twelve-requests-four-ways}.",
   "served": "the adapter was trained on 159 tracks from 41 releases § {#forty-one-releases-licence-by-licence}. the one listener, who was unblinded, stated that the model was not very good at making good edm and that most of the samples were not pieces of music they would ever listen to for pleasure § {#twelve-requests-four-ways}.",
   "seconds": 12.311198194000099,
   "ttft_ms": 959,
   "tokens_in": 12876,
   "tokens_out": 85,
   "cached_tokens": 32,
   "cites": [
    "forty-one-releases-licence-by-licence",
    "twelve-requests-four-ways"
   ],
   "gates": {
    "cites": [
     "forty-one-releases-licence-by-licence",
     "twelve-requests-four-ways"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [
    "159",
    "41"
   ],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "on-05",
   "kind": "on-page"
  },
  {
   "slug": "listen-for-yourself",
   "q": "how many renders are on the page, and what was done to their loudness?",
   "ok": true,
   "answer": "there are sixty-three players on the page, which play a listening master of each one § {#how-to-listen}. the mastered copies are created by applying one constant gain and nothing else to the raw render § {#how-to-listen}. the gain is the smaller of two numbers: how far the clip sits from −16 LUFS, and how far its true peak sits from a −1 dBTP ceiling § {#how-to-listen}. thirty-two of the clips land at −16 LUFS exactly, while the other thirty-one are left at their raw level because their peaks already sit at the engine's own −1 dB ceiling § {#how-to-listen}.",
   "served": "there are sixty-three players on the page, which play a listening master of each one § {#how-to-listen}. the mastered copies are created by applying one constant gain and nothing else to the raw render § {#how-to-listen}. the gain is the smaller of two numbers: how far the clip sits from −16 LUFS, and how far its true peak sits from a −1 dBTP ceiling § {#how-to-listen}. thirty-two of the clips land at −16 LUFS exactly, while the other thirty-one are left at their raw level because their peaks already sit at the engine's own −1 dB ceiling § {#how-to-listen}.",
   "seconds": 13.849587058999987,
   "ttft_ms": 1680,
   "tokens_in": 14146,
   "tokens_out": 145,
   "cached_tokens": 32,
   "cites": [
    "how-to-listen"
   ],
   "gates": {
    "cites": [
     "how-to-listen"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [
    "16",
    "16"
   ],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "on-06",
   "kind": "on-page"
  },
  {
   "slug": "half-an-hour-with-dead-composers",
   "q": "how long did the Sousa and Bach adapters take to train?",
   "ok": true,
   "answer": "the Sousa adapter took 14 minutes 36 seconds to train § {#sousa-a-hundred-and-six-marches}. the Bach adapter took 8 minutes 23 seconds § {#bach-the-pianist-who-freed-her-own-recordings}.",
   "served": "the Sousa adapter took 14 minutes 36 seconds to train § {#sousa-a-hundred-and-six-marches}. the Bach adapter took 8 minutes 23 seconds § {#bach-the-pianist-who-freed-her-own-recordings}.",
   "seconds": 1.4492857310001455,
   "ttft_ms": 357,
   "tokens_in": 5400,
   "tokens_out": 63,
   "cached_tokens": 0,
   "cites": [
    "sousa-a-hundred-and-six-marches",
    "bach-the-pianist-who-freed-her-own-recordings"
   ],
   "gates": {
    "cites": [
     "sousa-a-hundred-and-six-marches",
     "bach-the-pianist-who-freed-her-own-recordings"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [
    "14",
    "36",
    "8",
    "23"
   ],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "on-07",
   "kind": "on-page"
  },
  {
   "slug": "chopin-in-five-minutes",
   "q": "how many recordings were used, and how long did the training take by the trainer's clock?",
   "ok": true,
   "answer": "nineteen recordings went into training § {#five-minutes-of-training-honestly-accounted}. the training took 4 minutes 49 seconds by the trainer's own clock § {#five-minutes-of-training-honestly-accounted}.",
   "served": "nineteen recordings went into training § {#five-minutes-of-training-honestly-accounted}. the training took 4 minutes 49 seconds by the trainer's own clock § {#five-minutes-of-training-honestly-accounted}.",
   "seconds": 1.5726231070000267,
   "ttft_ms": 318,
   "tokens_in": 3550,
   "tokens_out": 56,
   "cached_tokens": 32,
   "cites": [
    "five-minutes-of-training-honestly-accounted"
   ],
   "gates": {
    "cites": [
     "five-minutes-of-training-honestly-accounted"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [
    "4",
    "49"
   ],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "on-08",
   "kind": "on-page"
  },
  {
   "slug": "the-same-sixteen",
   "q": "what did glm-5.3-flash score on the tool bench, and what did the local models score?",
   "ok": true,
   "answer": "the reference arm glm-5.3-flash:cloud scored 16/19 — 84.2%, Wilson 95% interval [62.4%, 94.5%] on the tool bench § {#sixteen-of-nineteen}. local models scored various amounts, including gemma4:26b (the seat) at 14/19 [51.2%, 88.2%] at think:false and 15/19 [56.7%, 91.5%] at think:true, qwen3.6:27b at 16/19 and 14/19, and gemma4:12b at 16/19 § {#sixteen-of-nineteen} § {#two-local-models-sixteen-days-earlier}.",
   "served": "the reference arm glm-5.3-flash:cloud scored 16/19 — 84.2%, Wilson 95% interval [62.4%, 94.5%] on the tool bench § {#sixteen-of-nineteen}. local models scored various amounts, including gemma4:26b (the seat) at 14/19 [51.2%, 88.2%] at think:false and 15/19 [56.7%, 91.5%] at think:true, qwen3.6:27b at 16/19 and 14/19, and gemma4:12b at 16/19 § {#sixteen-of-nineteen} § {#two-local-models-sixteen-days-earlier}.",
   "seconds": 5.979030255999987,
   "ttft_ms": 433,
   "tokens_in": 6233,
   "tokens_out": 189,
   "cached_tokens": 32,
   "cites": [
    "sixteen-of-nineteen",
    "two-local-models-sixteen-days-earlier"
   ],
   "gates": {
    "cites": [
     "sixteen-of-nineteen",
     "two-local-models-sixteen-days-earlier"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [
    "5.3",
    "16/19",
    "84.2",
    "95",
    "62.4",
    "94.5",
    "4:26",
    "14/19",
    "51.2",
    "88.2",
    "15/19",
    "56.7",
    "91.5",
    "3.6",
    "27",
    "16/19",
    "14/19",
    "4:12",
    "16/19"
   ],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "on-09",
   "kind": "on-page"
  },
  {
   "slug": "two-hours-on-battery",
   "q": "what speed did the mini pc hold on battery, and for how long?",
   "ok": false,
   "answer": "the mini pc held a speed of ~12 tokens/second on battery for about two hours § {#two-hours-at-12-tok-s-the-test-that-named-the-page}. specifically, the four answers completed on battery averaged 11.87 tok/s § {#two-hours-at-12-tok-s-the-test-that-named-the-page}.",
   "served": "this page doesn't say. the nearest it gets is § {#two-hours-at-12-tok-s-the-test-that-named-the-page}",
   "seconds": 4.0175912710001285,
   "ttft_ms": 714,
   "tokens_in": 10826,
   "tokens_out": 92,
   "cached_tokens": 32,
   "cites": [
    "two-hours-at-12-tok-s-the-test-that-named-the-page"
   ],
   "gates": {
    "cites": [
     "two-hours-at-12-tok-s-the-test-that-named-the-page"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": true,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [
    "12",
    "12",
    "11.87",
    "12"
   ],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "on-10",
   "kind": "on-page"
  },
  {
   "slug": "the-instrument-travels",
   "q": "when the frozen exam moved to the smaller card, where did the ranked model land?",
   "ok": true,
   "answer": "the qwen3.6:27b model, which was RANKED on the 96 GB card, was re-examined on a 24 GB consumer card and again earned a RANKED verdict § {#the-chair-and-who-gets-to-sit-in-it}. its judge trial scores on the smaller card were a kill-recall of 11 and a preservation of 8 § {#the-standings}.",
   "served": "the qwen3.6:27b model, which was RANKED on the 96 GB card, was re-examined on a 24 GB consumer card and again earned a RANKED verdict § {#the-chair-and-who-gets-to-sit-in-it}. its judge trial scores on the smaller card were a kill-recall of 11 and a preservation of 8 § {#the-standings}.",
   "seconds": 3.677064468000026,
   "ttft_ms": 650,
   "tokens_in": 8359,
   "tokens_out": 94,
   "cached_tokens": 32,
   "cites": [
    "the-chair-and-who-gets-to-sit-in-it",
    "the-standings"
   ],
   "gates": {
    "cites": [
     "the-chair-and-who-gets-to-sit-in-it",
     "the-standings"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [
    "3.6",
    "27",
    "96",
    "24",
    "11",
    "8"
   ],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "on-11",
   "kind": "on-page"
  },
  {
   "slug": "nine-worlds-one-dog",
   "q": "what stays the same across the nine paintings?",
   "ok": true,
   "answer": "the same large black Portuguese Water Dog with one bright white chest patch, the same curl, and the same tucked paws remain constant across all nine worlds § {#for-the-technically-curious}. the dog is also rendered taking the same nap under the same tree in every instance § {#for-the-technically-curious}.",
   "served": "the same large black Portuguese Water Dog with one bright white chest patch, the same curl, and the same tucked paws remain constant across all nine worlds § {#for-the-technically-curious}. the dog is also rendered taking the same nap under the same tree in every instance § {#for-the-technically-curious}.",
   "seconds": 3.277688051999803,
   "ttft_ms": 535,
   "tokens_in": 1889,
   "tokens_out": 70,
   "cached_tokens": 32,
   "cites": [
    "for-the-technically-curious"
   ],
   "gates": {
    "cites": [
     "for-the-technically-curious"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "on-12",
   "kind": "on-page"
  },
  {
   "slug": "what-150-watts-buys",
   "q": "what did the 450-watt cap cost the dense model, and what did it buy back?",
   "ok": true,
   "answer": "on the dense arm, the 600 → 450 W cap costs 2.0% of throughput and buys back 138 W, 11 °C, and a 28% jump in efficiency § {#the-answer-in-one-paragraph}. at concurrency 16, the throughput moved from 94.6 to 92.7 tok/s per stream § {#the-answer-in-one-paragraph}.",
   "served": "on the dense arm, the 600 → 450 W cap costs 2.0% of throughput and buys back 138 W, 11 °C, and a 28% jump in efficiency § {#the-answer-in-one-paragraph}. at concurrency 16, the throughput moved from 94.6 to 92.7 tok/s per stream § {#the-answer-in-one-paragraph}.",
   "seconds": 3.5960990290000154,
   "ttft_ms": 505,
   "tokens_in": 7036,
   "tokens_out": 100,
   "cached_tokens": 32,
   "cites": [
    "the-answer-in-one-paragraph"
   ],
   "gates": {
    "cites": [
     "the-answer-in-one-paragraph"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [
    "600",
    "450",
    "2.0",
    "138",
    "11",
    "28",
    "16",
    "94.6",
    "92.7"
   ],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "on-13",
   "kind": "on-page"
  },
  {
   "slug": "how-the-beat-lab-works",
   "q": "where is the sound made, and how long does a typical wish take?",
   "ok": true,
   "answer": "every sound in the Beat Lab is synthesised by the reader's own browser § {#the-sound-is-made-on-your-device}. a typical wish takes about 4 s § {#what-happens-when-you-ask-the-genie}.",
   "served": "every sound in the Beat Lab is synthesised by the reader's own browser § {#the-sound-is-made-on-your-device}. a typical wish takes about 4 s § {#what-happens-when-you-ask-the-genie}.",
   "seconds": 3.837351233000163,
   "ttft_ms": 2266,
   "tokens_in": 30921,
   "tokens_out": 59,
   "cached_tokens": 32,
   "cites": [
    "the-sound-is-made-on-your-device",
    "what-happens-when-you-ask-the-genie"
   ],
   "gates": {
    "cites": [
     "the-sound-is-made-on-your-device",
     "what-happens-when-you-ask-the-genie"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [
    "4"
   ],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "on-14",
   "kind": "on-page"
  },
  {
   "slug": "the-typist-and-the-developer",
   "q": "how long did the developer take to develop one picture?",
   "ok": true,
   "answer": "the developer developed a picture in 1.54 seconds when working alone § {#the-developer}. when the typist is also streaming tokens, the developer takes 1.82 seconds to develop a picture § {#one-card-both-jobs-at-once}.",
   "served": "the developer developed a picture in 1.54 seconds when working alone § {#the-developer}. when the typist is also streaming tokens, the developer takes 1.82 seconds to develop a picture § {#one-card-both-jobs-at-once}.",
   "seconds": 1.6652845899998283,
   "ttft_ms": 354,
   "tokens_in": 5363,
   "tokens_out": 58,
   "cached_tokens": 32,
   "cites": [
    "the-developer",
    "one-card-both-jobs-at-once"
   ],
   "gates": {
    "cites": [
     "the-developer",
     "one-card-both-jobs-at-once"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [
    "1.54",
    "1.82"
   ],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "on-15",
   "kind": "on-page"
  },
  {
   "slug": "a-rig-your-friend-already-owns",
   "q": "how much faster was the workstation card than the 3090 on the game's own art?",
   "ok": true,
   "answer": "the 96 GB card is about 3.4× the 3090 when using the FLUX.2 klein-4B sketch artist at the 512×512 live default resolution § {#the-sketch-ladder-the-number-the-game-lives-on}. for the 768×768 resolution, the 96 GB card is 3.4× faster than the 3090 § {#the-table-that-answers-the-question}. at 1024×1024, the 96 GB card is 3.4× the 3090 § {#the-sketch-ladder-the-number-the-game-lives-on}.",
   "served": "the 96 GB card is about 3.4× the 3090 when using the FLUX.2 klein-4B sketch artist at the 512×512 live default resolution § {#the-sketch-ladder-the-number-the-game-lives-on}. for the 768×768 resolution, the 96 GB card is 3.4× faster than the 3090 § {#the-table-that-answers-the-question}. at 1024×1024, the 96 GB card is 3.4× the 3090 § {#the-sketch-ladder-the-number-the-game-lives-on}.",
   "seconds": 3.3571599720000904,
   "ttft_ms": 340,
   "tokens_in": 4426,
   "tokens_out": 162,
   "cached_tokens": 32,
   "cites": [
    "the-sketch-ladder-the-number-the-game-lives-on",
    "the-table-that-answers-the-question"
   ],
   "gates": {
    "cites": [
     "the-sketch-ladder-the-number-the-game-lives-on",
     "the-table-that-answers-the-question"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [
    "96",
    "3.4",
    "3090",
    "2",
    "4",
    "512",
    "512",
    "768",
    "768",
    "96",
    "3.4",
    "3090",
    "1024",
    "1024",
    "96",
    "3.4",
    "3090"
   ],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "on-16",
   "kind": "on-page"
  },
  {
   "slug": "how-a-vision-model-sees",
   "q": "how many tokens does an 896 px picture become, and how long does its prefill take?",
   "ok": false,
   "answer": "on the seat the cove ships, an 896 px picture becomes 199 tokens and its prefill takes 184 ms § {#finding-one-a-picture-is-199-tokens-at-896-px-and-the}.",
   "served": "this page doesn't say. the nearest it gets is § {#finding-one-a-picture-is-199-tokens-at-896-px-and-the}",
   "seconds": 1.4745669760000055,
   "ttft_ms": 661,
   "tokens_in": 9992,
   "tokens_out": 61,
   "cached_tokens": 32,
   "cites": [
    "finding-one-a-picture-is-199-tokens-at-896-px-and-the"
   ],
   "gates": {
    "cites": [
     "finding-one-a-picture-is-199-tokens-at-896-px-and-the"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": true,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [
    "896",
    "199",
    "184",
    "199",
    "896"
   ],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "on-17",
   "kind": "on-page"
  },
  {
   "slug": "fifty-seven-milliseconds",
   "q": "which painter is the fastest cell, and at what time?",
   "ok": true,
   "answer": "the distilled painter is the sharpest and fastest cell in the whole second run, recorded at 0.62 s at 4 steps § {#finding-four-the-argument-with-our-own-settings-and-what-it}.",
   "served": "the distilled painter is the sharpest and fastest cell in the whole second run, recorded at 0.62 s at 4 steps § {#finding-four-the-argument-with-our-own-settings-and-what-it}.",
   "seconds": 1.1161941259999821,
   "ttft_ms": 523,
   "tokens_in": 5191,
   "tokens_out": 53,
   "cached_tokens": 32,
   "cites": [
    "finding-four-the-argument-with-our-own-settings-and-what-it"
   ],
   "gates": {
    "cites": [
     "finding-four-the-argument-with-our-own-settings-and-what-it"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [
    "0.62",
    "4"
   ],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "on-18",
   "kind": "on-page"
  },
  {
   "slug": "the-dog-the-dice-and-the-painter",
   "q": "how big is the file the whole world is written in?",
   "ok": true,
   "answer": "the world of RealKeep is written in one YAML file that contains 16,705 lines § {#the-world-is-a-file}. this file holds six regions, forty NPCs, fifty-nine quests, and three kinds of adoptable pet § {#the-world-is-a-file}.",
   "served": "the world of RealKeep is written in one YAML file that contains 16,705 lines § {#the-world-is-a-file}. this file holds six regions, forty NPCs, fifty-nine quests, and three kinds of adoptable pet § {#the-world-is-a-file}.",
   "seconds": 1.9199443810000503,
   "ttft_ms": 416,
   "tokens_in": 5873,
   "tokens_out": 67,
   "cached_tokens": 32,
   "cites": [
    "the-world-is-a-file"
   ],
   "gates": {
    "cites": [
     "the-world-is-a-file"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [
    "16,705"
   ],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "on-19",
   "kind": "on-page"
  },
  {
   "slug": "where-new-knowledge-comes-from",
   "q": "what are the two engines the page names for where new knowledge comes from?",
   "ok": true,
   "answer": "the two engines named are present in the data, unmined, which refers to an archive that was already there but lacked a wide enough reader § {#the-first-engine-the-archive-nobody-had-read}, and implied by the rules, unvisited, which refers to a rule captured well enough that it emits entries no row of the data contains § {#the-second-engine-the-rooms-the-rules-imply}. these engines represent the two kinds of there-all-along § {#two-kinds-of-there-all-along}.",
   "served": "the two engines named are present in the data, unmined, which refers to an archive that was already there but lacked a wide enough reader § {#the-first-engine-the-archive-nobody-had-read}, and implied by the rules, unvisited, which refers to a rule captured well enough that it emits entries no row of the data contains § {#the-second-engine-the-rooms-the-rules-imply}. these engines represent the two kinds of there-all-along § {#two-kinds-of-there-all-along}.",
   "seconds": 3.469866613000022,
   "ttft_ms": 453,
   "tokens_in": 5220,
   "tokens_out": 120,
   "cached_tokens": 32,
   "cites": [
    "the-first-engine-the-archive-nobody-had-read",
    "the-second-engine-the-rooms-the-rules-imply",
    "two-kinds-of-there-all-along"
   ],
   "gates": {
    "cites": [
     "the-first-engine-the-archive-nobody-had-read",
     "the-second-engine-the-rooms-the-rules-imply",
     "two-kinds-of-there-all-along"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "on-20",
   "kind": "on-page"
  },
  {
   "slug": "compression-is-the-objective",
   "q": "what does the page say a language model is, and what prize does it point to?",
   "ok": true,
   "answer": "the article states that a model's weights are the compression, acting as the codebook for a corpus § {#the-punchline}. it also notes that training a large language model involves minimizing surprise by the next token of its training text across trillions of them § {#the-punchline}. the text points to the Hutter Prize, which charges for everything including every decompressor byte counted § {#what-we-can-show-you}.",
   "served": "the article states that a model's weights are the compression, acting as the codebook for a corpus § {#the-punchline}. it also notes that training a large language model involves minimizing surprise by the next token of its training text across trillions of them § {#the-punchline}. the text points to the Hutter Prize, which charges for everything including every decompressor byte counted § {#what-we-can-show-you}.",
   "seconds": 2.2599917270001697,
   "ttft_ms": 565,
   "tokens_in": 5230,
   "tokens_out": 92,
   "cached_tokens": 32,
   "cites": [
    "the-punchline",
    "what-we-can-show-you"
   ],
   "gates": {
    "cites": [
     "the-punchline",
     "what-we-can-show-you"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "on-21",
   "kind": "on-page"
  },
  {
   "slug": "the-free-speed-wasnt-free",
   "q": "what was the production median before the change, and what caused the speed-up?",
   "ok": true,
   "answer": "the production median was 138.6 tok/s over 1,982 live rulings on 2026-08-15 § {#the-day-the-box-got-faster}. the speed-up was caused by an ollama upgrade that changed the default repeat_penalty from 1.1 to 1.0 for models that do not set one § {#the-two-line-commit}. this change allowed the server to skip the penalties stage in the sampling pipeline § {#the-hunt-change-one-thing}.",
   "served": "the production median was 138.6 tok/s over 1,982 live rulings on 2026-08-15 § {#the-day-the-box-got-faster}. the speed-up was caused by an ollama upgrade that changed the default repeat_penalty from 1.1 to 1.0 for models that do not set one § {#the-two-line-commit}. this change allowed the server to skip the penalties stage in the sampling pipeline § {#the-hunt-change-one-thing}.",
   "seconds": 3.3999083110002175,
   "ttft_ms": 754,
   "tokens_in": 7223,
   "tokens_out": 120,
   "cached_tokens": 32,
   "cites": [
    "the-day-the-box-got-faster",
    "the-two-line-commit",
    "the-hunt-change-one-thing"
   ],
   "gates": {
    "cites": [
     "the-day-the-box-got-faster",
     "the-two-line-commit",
     "the-hunt-change-one-thing"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [
    "138.6",
    "1,982",
    "2026-08-15",
    "1.1",
    "1.0"
   ],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "on-22",
   "kind": "on-page"
  },
  {
   "slug": "the-compressed-photograph",
   "q": "how much slower was the uncompressed original than the compressed build?",
   "ok": true,
   "answer": "the uncompressed original was 3.4× the speed of the trained-4-bit build § {#the-same-question-down-the-whole-ladder}. specifically, the uncompressed original's speed was 58.1 tok/s compared to the trained-4-bit build's 134.4 tok/s § {#the-same-question-down-the-whole-ladder}.",
   "served": "the uncompressed original was 3.4× the speed of the trained-4-bit build § {#the-same-question-down-the-whole-ladder}. specifically, the uncompressed original's speed was 58.1 tok/s compared to the trained-4-bit build's 134.4 tok/s § {#the-same-question-down-the-whole-ladder}.",
   "seconds": 2.3108695490000173,
   "ttft_ms": 1024,
   "tokens_in": 8205,
   "tokens_out": 91,
   "cached_tokens": 32,
   "cites": [
    "the-same-question-down-the-whole-ladder"
   ],
   "gates": {
    "cites": [
     "the-same-question-down-the-whole-ladder"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [
    "3.4",
    "4",
    "58.1",
    "4",
    "134.4"
   ],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "on-23",
   "kind": "on-page"
  },
  {
   "slug": "three-at-the-table",
   "q": "what is the true active share for the live seat, and what does the page call the two numbers in a model name?",
   "ok": true,
   "answer": "the true active share for the live seat is 14.4% of the model's bytes § {#the-payroll-and-the-meeting}. the two numbers in a model name are the payroll, which represents the parameters hired, trained, and taking up space, and the meeting, which represents the parameters that actually show up for any given token § {#the-tag-with-two-numbers}.",
   "served": "the true active share for the live seat is 14.4% of the model's bytes § {#the-payroll-and-the-meeting}. the two numbers in a model name are the payroll, which represents the parameters hired, trained, and taking up space, and the meeting, which represents the parameters that actually show up for any given token § {#the-tag-with-two-numbers}.",
   "seconds": 8.588107052000169,
   "ttft_ms": 447,
   "tokens_in": 7404,
   "tokens_out": 86,
   "cached_tokens": 32,
   "cites": [
    "the-payroll-and-the-meeting",
    "the-tag-with-two-numbers"
   ],
   "gates": {
    "cites": [
     "the-payroll-and-the-meeting",
     "the-tag-with-two-numbers"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [
    "14.4"
   ],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "on-24",
   "kind": "on-page"
  },
  {
   "slug": "reading-is-fast",
   "q": "what was the median reading speed, and why is reading faster than writing?",
   "ok": true,
   "answer": "the median reading speed was 5,415 tokens/s § {#a-measurement-from-our-own-card}. reading is faster because it is parallel, as the model can process every token in the prompt at the same time since the text already exists § {#reading-is-a-group-photo}. in contrast, writing is serial because the future of the sentence does not yet exist, requiring the model to choose tokens one at a time § {#writing-is-one-at-a-time}.",
   "served": "the median reading speed was 5,415 tokens/s § {#a-measurement-from-our-own-card}. reading is faster because it is parallel, as the model can process every token in the prompt at the same time since the text already exists § {#reading-is-a-group-photo}. in contrast, writing is serial because the future of the sentence does not yet exist, requiring the model to choose tokens one at a time § {#writing-is-one-at-a-time}.",
   "seconds": 8.747530811999923,
   "ttft_ms": 485,
   "tokens_in": 5262,
   "tokens_out": 109,
   "cached_tokens": 32,
   "cites": [
    "a-measurement-from-our-own-card",
    "reading-is-a-group-photo",
    "writing-is-one-at-a-time"
   ],
   "gates": {
    "cites": [
     "a-measurement-from-our-own-card",
     "reading-is-a-group-photo",
     "writing-is-one-at-a-time"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [
    "5,415"
   ],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "on-25",
   "kind": "on-page"
  },
  {
   "slug": "three-librarians",
   "q": "how many dimensions does the embedder's map have, and what does the third librarian do?",
   "ok": true,
   "answer": "the embedder's map has 768 directions § {#the-second-librarian-searches-by-meaning}. the third librarian works only the aisle of the specific game being asked about, picking up the edition's own rulebook and any matched correction documents § {#the-third-librarian-never-left-your-aisle}. she ranks each book's passages by the map of meaning and sends up to thirty-two candidates into the room § {#the-third-librarian-never-left-your-aisle}.",
   "served": "the embedder's map has 768 directions § {#the-second-librarian-searches-by-meaning}. the third librarian works only the aisle of the specific game being asked about, picking up the edition's own rulebook and any matched correction documents § {#the-third-librarian-never-left-your-aisle}. she ranks each book's passages by the map of meaning and sends up to thirty-two candidates into the room § {#the-third-librarian-never-left-your-aisle}.",
   "seconds": 8.99833480899997,
   "ttft_ms": 383,
   "tokens_in": 5643,
   "tokens_out": 117,
   "cached_tokens": 32,
   "cites": [
    "the-second-librarian-searches-by-meaning",
    "the-third-librarian-never-left-your-aisle"
   ],
   "gates": {
    "cites": [
     "the-second-librarian-searches-by-meaning",
     "the-third-librarian-never-left-your-aisle"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [
    "768"
   ],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "on-26",
   "kind": "on-page"
  },
  {
   "slug": "the-new-kid",
   "q": "which seat did the new model pass, and which floor did it miss?",
   "ok": true,
   "answer": "the qwen3.8:27b model passed the classifier seat, clearing all five pre-registered gates § {#the-classifier-seat-five-gates-all-pass-faster}. however, it failed the judge seat by missing the preservation floor of 13, scoring 11 of 16 § {#the-judge-seat-and-the-floor-it-missed}.",
   "served": "the qwen3.8:27b model passed the classifier seat, clearing all five pre-registered gates § {#the-classifier-seat-five-gates-all-pass-faster}. however, it failed the judge seat by missing the preservation floor of 13, scoring 11 of 16 § {#the-judge-seat-and-the-floor-it-missed}.",
   "seconds": 8.143793357999812,
   "ttft_ms": 930,
   "tokens_in": 14813,
   "tokens_out": 86,
   "cached_tokens": 32,
   "cites": [
    "the-classifier-seat-five-gates-all-pass-faster",
    "the-judge-seat-and-the-floor-it-missed"
   ],
   "gates": {
    "cites": [
     "the-classifier-seat-five-gates-all-pass-faster",
     "the-judge-seat-and-the-floor-it-missed"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [
    "3.8",
    "27",
    "13",
    "11",
    "16"
   ],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "on-27",
   "kind": "on-page"
  },
  {
   "slug": "llms-txt",
   "q": "did any ai crawler fetch the file in the first thirty days, and how big is the largest measured llms-full file?",
   "ok": true,
   "answer": "the article states that in the original twenty-six days, there were zero requests from an AI crawler § {#measured-again-after-publication}. the largest llms-full file measured weighs 30.7 MiB § {#where-the-file-came-from}.",
   "served": "the article states that in the original twenty-six days, there were zero requests from an AI crawler § {#measured-again-after-publication}. the largest llms-full file measured weighs 30.7 MiB § {#where-the-file-came-from}.",
   "seconds": 7.840764402000104,
   "ttft_ms": 780,
   "tokens_in": 12220,
   "tokens_out": 59,
   "cached_tokens": 32,
   "cites": [
    "measured-again-after-publication",
    "where-the-file-came-from"
   ],
   "gates": {
    "cites": [
     "measured-again-after-publication",
     "where-the-file-came-from"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [
    "30.7"
   ],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "on-28",
   "kind": "on-page"
  },
  {
   "slug": "the-open-call",
   "q": "how many models were handed the frozen moment, and what did the round cost?",
   "ok": true,
   "answer": "twenty arms were seated to answer the frozen bytes § {#five-things-before-any-figure}. the metered spend for the round was $2.2826 § {$2.2826, over 56 metered calls, against $38.49 of caps}.",
   "served": "twenty arms were seated to answer the frozen bytes § {#five-things-before-any-figure}. the metered spend for the round was $2.2826 § {$2.2826, over 56 metered calls, against $38.49 of caps}.",
   "seconds": 8.232999726999878,
   "ttft_ms": 6734,
   "tokens_in": 63328,
   "tokens_out": 64,
   "cached_tokens": 32,
   "cites": [
    "five-things-before-any-figure"
   ],
   "gates": {
    "cites": [
     "five-things-before-any-figure"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [
    "2.2826",
    "2.2826",
    "56",
    "38.49"
   ],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "on-29",
   "kind": "on-page"
  },
  {
   "slug": "where-your-question-goes",
   "q": "how long does every notebook on the server keep its entries?",
   "ok": false,
   "answer": "every notebook on the server now keeps thirty days instead of a century § {#the-first-promise-your-visit-stays-between-us}. a sweep runs every night to hold this thirty days § {#the-first-promise-your-visit-stays-between-us}.",
   "served": "this page doesn't say. the nearest it gets is § {#what-to-take-with-you}",
   "seconds": 1.25812224099991,
   "ttft_ms": 312,
   "tokens_in": 4240,
   "tokens_out": 60,
   "cached_tokens": 32,
   "cites": [
    "the-first-promise-your-visit-stays-between-us"
   ],
   "gates": {
    "cites": [
     "the-first-promise-your-visit-stays-between-us"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": true,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [],
   "figures_verbatim": true,
   "cites_valid": false,
   "redaction_hits": 0,
   "id": "on-30",
   "kind": "on-page"
  },
  {
   "slug": "how-we-work",
   "q": "what does the page say about the humans and agents who make each exhibit?",
   "ok": false,
   "answer": "the human generates and judges while the agent amplifies and verifies § {#what-to-take-with-you}. agents perform tasks such as recon, planning, and drafting, while the human operator provides the noticing, the roster conversation, the hardening panel review, and the final release § {#the-thorough-walk}. the human makes the work true by deciding what is worth testing and what ships, whereas agents make the work bigger by testing more models and verifying more claims § {#the-thorough-walk}.",
   "served": "this page doesn't say. the nearest it gets is § {#what-to-take-with-you}",
   "seconds": 4.821453788000099,
   "ttft_ms": 338,
   "tokens_in": 4657,
   "tokens_out": 106,
   "cached_tokens": 32,
   "cites": [
    "what-to-take-with-you",
    "the-thorough-walk"
   ],
   "gates": {
    "cites": [
     "what-to-take-with-you",
     "the-thorough-walk"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": true,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [],
   "figures_verbatim": true,
   "cites_valid": false,
   "redaction_hits": 0,
   "id": "on-31",
   "kind": "on-page"
  },
  {
   "slug": "the-move",
   "q": "what did the median search time go from and to after the move?",
   "ok": true,
   "answer": "the median search time went from 242 ms across five weeks of laptop-era answers to 77.5 ms in the new era § {#what-to-take-with-you}. this comparison is described as the retrieval leg the move actually touched § {#what-to-take-with-you}.",
   "served": "the median search time went from 242 ms across five weeks of laptop-era answers to 77.5 ms in the new era § {#what-to-take-with-you}. this comparison is described as the retrieval leg the move actually touched § {#what-to-take-with-you}.",
   "seconds": 1.1534605809999903,
   "ttft_ms": 501,
   "tokens_in": 5932,
   "tokens_out": 67,
   "cached_tokens": 32,
   "cites": [
    "what-to-take-with-you"
   ],
   "gates": {
    "cites": [
     "what-to-take-with-you"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [
    "242",
    "77.5"
   ],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "on-32",
   "kind": "on-page"
  },
  {
   "slug": "chair-trials",
   "q": "how many models and arms sat the five exams, and did the box stop serving?",
   "ok": false,
   "answer": "nine models and eleven arms were part of the roster, though only five of the eleven arms sat all five chairs § {#the-roster, and what it takes to load them}. the serving probes show that the box did not stop serving, as twenty-six real asks were registered, sent and served § {#the serving probes}.",
   "served": "this page doesn't say. the nearest it gets is § {#what-to-take-with-you}",
   "seconds": 5.193013212000096,
   "ttft_ms": 2943,
   "tokens_in": 29864,
   "tokens_out": 68,
   "cached_tokens": 32,
   "cites": [],
   "gates": {
    "cites": [],
    "redacted": false,
    "figure_unverified": false,
    "uncited": true,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [],
   "figures_verbatim": true,
   "cites_valid": false,
   "redaction_hits": 0,
   "id": "on-33",
   "kind": "on-page"
  },
  {
   "slug": "cove-voice-head-to-head",
   "q": "what happened to the August class's best voice at the cove's gate?",
   "ok": false,
   "answer": "the article does not state what happened to the August class's best voice at the cove's gate § {#floor-1---canon}. the text only mentions that the panel's warmest voice in the August arrivals was a model that invented names for the drowned diver twice § {#floor-1---canon}.",
   "served": "this page doesn't say. the nearest it gets is § {#what-to-take-with-you}",
   "seconds": 4.908914791999905,
   "ttft_ms": 1045,
   "tokens_in": 16187,
   "tokens_out": 64,
   "cached_tokens": 32,
   "cites": [
    "floor-1---canon"
   ],
   "gates": {
    "cites": [
     "floor-1---canon"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": true,
    "new_noun": true,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [],
   "figures_verbatim": true,
   "cites_valid": false,
   "redaction_hits": 0,
   "id": "on-34",
   "kind": "on-page"
  },
  {
   "slug": "outside-judges",
   "q": "how many outside labs re-judged, and what was the widest cross-vendor gap?",
   "ok": false,
   "answer": "four rival labs re-judged the sealed rounds § {#what-to-take-with-you}. the widest cross-vendor gap on any arm is 0.178 § {#the-honest-disagreements-print-too}.",
   "served": "this page doesn't say. the nearest it gets is § {#what-to-take-with-you}",
   "seconds": 4.62653547900004,
   "ttft_ms": 3019,
   "tokens_in": 9946,
   "tokens_out": 53,
   "cached_tokens": 32,
   "cites": [
    "what-to-take-with-you",
    "the-honest-disagreements-print-too"
   ],
   "gates": {
    "cites": [
     "what-to-take-with-you",
     "the-honest-disagreements-print-too"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": true,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [
    "0.178"
   ],
   "figures_verbatim": true,
   "cites_valid": false,
   "redaction_hits": 0,
   "id": "on-35",
   "kind": "on-page"
  },
  {
   "slug": "august-arrivals",
   "q": "how many models landed that week, and what share broke the response contract?",
   "ok": true,
   "answer": "four fresh models arrived that week, and all four broke the response contract on more than 10% of calls § {#what-to-take-with-you}. specifically, the failure rates for the cohort were 16.3%, 51.2%, 55.8% and 62.8% of 129 § {#what-to-take-with-you}.",
   "served": "four fresh models arrived that week, and all four broke the response contract on more than 10% of calls § {#what-to-take-with-you}. specifically, the failure rates for the cohort were 16.3%, 51.2%, 55.8% and 62.8% of 129 § {#what-to-take-with-you}.",
   "seconds": 4.77852650900013,
   "ttft_ms": 3470,
   "tokens_in": 10841,
   "tokens_out": 87,
   "cached_tokens": 32,
   "cites": [
    "what-to-take-with-you"
   ],
   "gates": {
    "cites": [
     "what-to-take-with-you"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [
    "10",
    "16.3",
    "51.2",
    "55.8",
    "62.8",
    "129"
   ],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "on-36",
   "kind": "on-page"
  },
  {
   "slug": "seat-trials",
   "q": "what are the kill-recall and preservation floors a model must clear?",
   "ok": false,
   "answer": "the kill-recall floor is ≥ 23/27 and the preservation floor is ≥ 13/16 § {#the-instrument}. these thresholds are applied to 43 claim-cases at three repeats each at temperature zero § {#the-instrument}.",
   "served": "this page doesn't say. the nearest it gets is § {#what-to-take-with-you}",
   "seconds": 4.405161866000071,
   "ttft_ms": 3376,
   "tokens_in": 6861,
   "tokens_out": 56,
   "cached_tokens": 32,
   "cites": [
    "the-instrument"
   ],
   "gates": {
    "cites": [
     "the-instrument"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": true,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [
    "23/27",
    "13/16",
    "43"
   ],
   "figures_verbatim": true,
   "cites_valid": false,
   "redaction_hits": 0,
   "id": "on-37",
   "kind": "on-page"
  },
  {
   "slug": "voice-trials",
   "q": "how many model tags were tried, and what does the anchor model read?",
   "ok": false,
   "answer": "the article states that there were 24 distinct model tags across 40 scored runs § {#provenance}. the anchor model reads 9.0, 7.0 and 6.2 on three record sets, unchanged § {#what-to-take-with-you}.",
   "served": "this page doesn't say. the nearest it gets is § {#what-to-take-with-you}",
   "seconds": 1.2307228199999827,
   "ttft_ms": 589,
   "tokens_in": 9943,
   "tokens_out": 60,
   "cached_tokens": 32,
   "cites": [
    "provenance",
    "what-to-take-with-you"
   ],
   "gates": {
    "cites": [
     "provenance",
     "what-to-take-with-you"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": true,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [
    "24",
    "40",
    "9.0",
    "7.0",
    "6.2"
   ],
   "figures_verbatim": true,
   "cites_valid": false,
   "redaction_hits": 0,
   "id": "on-38",
   "kind": "on-page"
  },
  {
   "slug": "licences",
   "q": "what does the page say it does with each model's licence?",
   "ok": false,
   "answer": "the page states that every dated row represents a licence text that was read in full by the workshop agents § {#provenance}. the method for each read is named once per table so that a reader can repeat the process themselves § {#how-to-read-this-page}. the ledger is re-read whenever a seat changes or a model updates, and the workshop sets a floor to ensure every row gets a fresh read at least every six months § {#why-this-page-is-called-living}.",
   "served": "this page doesn't say. the nearest it gets is § The licence ledger — every model, read first-hand",
   "seconds": 13.297148950000064,
   "ttft_ms": 749,
   "tokens_in": 8189,
   "tokens_out": 105,
   "cached_tokens": 32,
   "cites": [
    "provenance",
    "how-to-read-this-page",
    "why-this-page-is-called-living"
   ],
   "gates": {
    "cites": [
     "provenance",
     "how-to-read-this-page",
     "why-this-page-is-called-living"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": true,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [],
   "figures_verbatim": true,
   "cites_valid": false,
   "redaction_hits": 0,
   "id": "on-39",
   "kind": "on-page"
  },
  {
   "slug": "what-150-watts-buys",
   "q": "what does a large cloud provider charge for the same work?",
   "ok": true,
   "answer": "the article does not answer this question § {#the-dial-and-who-notices}.",
   "served": "the article does not answer this question § {#the-dial-and-who-notices}.",
   "seconds": 11.411550268999918,
   "ttft_ms": 451,
   "tokens_in": 7026,
   "tokens_out": 21,
   "cached_tokens": 32,
   "cites": [
    "the-dial-and-who-notices"
   ],
   "gates": {
    "cites": [
     "the-dial-and-who-notices"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "off-01",
   "kind": "off-page"
  },
  {
   "slug": "how-we-work",
   "q": "what is the postal address of the workshop that ran this?",
   "ok": false,
   "answer": "the article does not provide a postal address for the workshop § {#the-honest-answer}.",
   "served": "this page doesn't say. the nearest it gets is § {#the-velocity-receipt-is-this-page}",
   "seconds": 11.315809403000003,
   "ttft_ms": 821,
   "tokens_in": 4654,
   "tokens_out": 20,
   "cached_tokens": 32,
   "cites": [
    "the-honest-answer"
   ],
   "gates": {
    "cites": [
     "the-honest-answer"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": true,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [],
   "figures_verbatim": true,
   "cites_valid": false,
   "redaction_hits": 0,
   "id": "off-02",
   "kind": "off-page"
  },
  {
   "slug": "llms-txt",
   "q": "how many people subscribe to this site's feed?",
   "ok": true,
   "answer": "the article does not answer how many people subscribe to this site's feed, though it discusses request counts for llms.txt files in § {#does-anyone-fetch-it}.",
   "served": "the article does not answer how many people subscribe to this site's feed, though it discusses request counts for llms.txt files in § {#does-anyone-fetch-it}.",
   "seconds": 11.407042696999952,
   "ttft_ms": 1493,
   "tokens_in": 12205,
   "tokens_out": 39,
   "cached_tokens": 32,
   "cites": [
    "does-anyone-fetch-it"
   ],
   "gates": {
    "cites": [
     "does-anyone-fetch-it"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "off-03",
   "kind": "off-page"
  },
  {
   "slug": "three-new-voices-at-the-narrators-chair",
   "q": "what will the next article on this site be about?",
   "ok": true,
   "answer": "the article does not state what the next article on the site will be about § {#the-rest-of-the-seminar}.",
   "served": "the article does not state what the next article on the site will be about § {#the-rest-of-the-seminar}.",
   "seconds": 11.624769480000168,
   "ttft_ms": 10946,
   "tokens_in": 82391,
   "tokens_out": 29,
   "cached_tokens": 32,
   "cites": [
    "the-rest-of-the-seminar"
   ],
   "gates": {
    "cites": [
     "the-rest-of-the-seminar"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "off-04",
   "kind": "off-page"
  },
  {
   "slug": "two-hours-on-battery",
   "q": "what did this cost in electricity at my own utility's rate?",
   "ok": false,
   "answer": "the article does not answer what this cost in electricity at the reader's own utility's rate § {#two-hours-at-12-tok-s-the-test-that-named-the-page}.",
   "served": "this page doesn't say. the nearest it gets is § {#a-few-words-this-page-leans-on}",
   "seconds": 11.640782983999998,
   "ttft_ms": 1151,
   "tokens_in": 10825,
   "tokens_out": 48,
   "cached_tokens": 32,
   "cites": [
    "two-hours-at-12-tok-s-the-test-that-named-the-page"
   ],
   "gates": {
    "cites": [
     "two-hours-at-12-tok-s-the-test-that-named-the-page"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": true,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [
    "12"
   ],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "off-05",
   "kind": "off-page"
  },
  {
   "slug": "reading-is-fast",
   "q": "who won the football world cup in 2022?",
   "ok": true,
   "answer": "the article does not answer this question § {#the-mystery-stated-honestly}.",
   "served": "the article does not answer this question § {#the-mystery-stated-honestly}.",
   "seconds": 0.7081923240000378,
   "ttft_ms": 395,
   "tokens_in": 5260,
   "tokens_out": 19,
   "cached_tokens": 0,
   "cites": [
    "the-mystery-stated-honestly"
   ],
   "gates": {
    "cites": [
     "the-mystery-stated-honestly"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "off-06",
   "kind": "off-page"
  },
  {
   "slug": "a-rig-your-friend-already-owns",
   "q": "what does one of these cards cost in my country today?",
   "ok": true,
   "answer": "the article does not answer the question about the cost of the cards in the reader's country today § {#the-moment-this-page-is-about}.",
   "served": "the article does not answer the question about the cost of the cards in the reader's country today § {#the-moment-this-page-is-about}.",
   "seconds": 1.5443928050001432,
   "ttft_ms": 481,
   "tokens_in": 4416,
   "tokens_out": 35,
   "cached_tokens": 32,
   "cites": [
    "the-moment-this-page-is-about"
   ],
   "gates": {
    "cites": [
     "the-moment-this-page-is-about"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "off-07",
   "kind": "off-page"
  },
  {
   "slug": "the-move",
   "q": "can you write a python script that reproduces this for me?",
   "ok": false,
   "answer": "the article does not provide a python script or instructions for writing one, though it mentions that the bench instrument is a frozen runner executed in place and unmodified § {#instrument one · the bench, re-run}. the text provides links to machine-readable files such as c6-vps-rerun.json and receipts-era-stats.json § {#how to check our work — and see it live}.",
   "served": "this page doesn't say. the nearest it gets is § {#what-to-take-with-you}",
   "seconds": 8.918386601999828,
   "ttft_ms": 447,
   "tokens_in": 5930,
   "tokens_out": 85,
   "cached_tokens": 32,
   "cites": [],
   "gates": {
    "cites": [],
    "redacted": false,
    "figure_unverified": false,
    "uncited": true,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [
    "6"
   ],
   "figures_verbatim": true,
   "cites_valid": false,
   "redaction_hits": 0,
   "id": "off-08",
   "kind": "off-page"
  },
  {
   "slug": "fifty-seven-milliseconds",
   "q": "what was the weather like on the day this was measured?",
   "ok": true,
   "answer": "the article does not answer the question about the weather § {#how-to-check-our-work-and-see-it-live}.",
   "served": "the article does not answer the question about the weather § {#how-to-check-our-work-and-see-it-live}.",
   "seconds": 1.2035798399999749,
   "ttft_ms": 498,
   "tokens_in": 5191,
   "tokens_out": 31,
   "cached_tokens": 32,
   "cites": [
    "how-to-check-our-work-and-see-it-live"
   ],
   "gates": {
    "cites": [
     "how-to-check-our-work-and-see-it-live"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "off-09",
   "kind": "off-page"
  },
  {
   "slug": "six-worlds-painted-large",
   "q": "how many hours a week do the people who ran this work?",
   "ok": true,
   "answer": "the article does not say how many hours a week the people who ran this work § {#six-worlds-for-a-print-lab-that-is-coming-soon}.",
   "served": "the article does not say how many hours a week the people who ran this work § {#six-worlds-for-a-print-lab-that-is-coming-soon}.",
   "seconds": 1.3161832309999681,
   "ttft_ms": 407,
   "tokens_in": 2604,
   "tokens_out": 39,
   "cached_tokens": 32,
   "cites": [
    "six-worlds-for-a-print-lab-that-is-coming-soon"
   ],
   "gates": {
    "cites": [
     "six-worlds-for-a-print-lab-that-is-coming-soon"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "off-10",
   "kind": "off-page"
  },
  {
   "slug": "outside-judges",
   "q": "what is this workshop's hourly rate for consulting work?",
   "ok": true,
   "answer": "the article does not answer the question about the workshop's hourly rate for consulting work § {#what-to-take-with-you}.",
   "served": "the article does not answer the question about the workshop's hourly rate for consulting work § {#what-to-take-with-you}.",
   "seconds": 1.2298247149999497,
   "ttft_ms": 751,
   "tokens_in": 9939,
   "tokens_out": 30,
   "cached_tokens": 32,
   "cites": [
    "what-to-take-with-you"
   ],
   "gates": {
    "cites": [
     "what-to-take-with-you"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "off-11",
   "kind": "off-page"
  },
  {
   "slug": "the-typist-and-the-developer",
   "q": "what is this workshop's annual revenue?",
   "ok": true,
   "answer": "the article does not answer the question about the workshop's annual revenue § {#one-card-both-jobs-at-once}.",
   "served": "the article does not answer the question about the workshop's annual revenue § {#one-card-both-jobs-at-once}.",
   "seconds": 2.316660924000189,
   "ttft_ms": 379,
   "tokens_in": 5361,
   "tokens_out": 29,
   "cached_tokens": 32,
   "cites": [
    "one-card-both-jobs-at-once"
   ],
   "gates": {
    "cites": [
     "one-card-both-jobs-at-once"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "off-12",
   "kind": "off-page"
  },
  {
   "slug": "compression-is-the-objective",
   "q": "how would these numbers change on windows instead?",
   "ok": true,
   "answer": "the article does not say how these numbers would change on windows instead § {#how-to-check-our-work-and-see-it-live}.",
   "served": "the article does not say how these numbers would change on windows instead § {#how-to-check-our-work-and-see-it-live}.",
   "seconds": 7.639896845999829,
   "ttft_ms": 374,
   "tokens_in": 5221,
   "tokens_out": 34,
   "cached_tokens": 32,
   "cites": [
    "how-to-check-our-work-and-see-it-live"
   ],
   "gates": {
    "cites": [
     "how-to-check-our-work-and-see-it-live"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "off-13",
   "kind": "off-page"
  },
  {
   "slug": "the-open-call",
   "q": "what did other reviewers say about this piece after it was published?",
   "ok": true,
   "answer": "the article does not state what other reviewers said about this piece after it was published § {#the-words-this-page-leans-on}.",
   "served": "the article does not state what other reviewers said about this piece after it was published § {#the-words-this-page-leans-on}.",
   "seconds": 7.871456843000033,
   "ttft_ms": 7178,
   "tokens_in": 63325,
   "tokens_out": 31,
   "cached_tokens": 32,
   "cites": [
    "the-words-this-page-leans-on"
   ],
   "gates": {
    "cites": [
     "the-words-this-page-leans-on"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "off-14",
   "kind": "off-page"
  },
  {
   "slug": "the-new-kid",
   "q": "how does this compare to a result published in 2028?",
   "ok": false,
   "answer": "the article does not provide information regarding any results published in 2028 § {#the-arrival}.",
   "served": "this page doesn't say. the nearest it gets is § {#the-arrival}",
   "seconds": 7.379917046999935,
   "ttft_ms": 1041,
   "tokens_in": 14813,
   "tokens_out": 23,
   "cached_tokens": 32,
   "cites": [
    "the-arrival"
   ],
   "gates": {
    "cites": [
     "the-arrival"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": true,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "off-15",
   "kind": "off-page"
  },
  {
   "slug": "where-your-question-goes",
   "q": "which country is the workshop's business registered in?",
   "ok": true,
   "answer": "the article does not answer the question about which country the workshop's business is registered in § {#what-to-take-with-you}.",
   "served": "the article does not answer the question about which country the workshop's business is registered in § {#what-to-take-with-you}.",
   "seconds": 7.2870014529999025,
   "ttft_ms": 328,
   "tokens_in": 4239,
   "tokens_out": 31,
   "cached_tokens": 32,
   "cites": [
    "what-to-take-with-you"
   ],
   "gates": {
    "cites": [
     "what-to-take-with-you"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "off-16",
   "kind": "off-page"
  },
  {
   "slug": "nine-worlds-one-dog",
   "q": "how many other readers have asked this same question?",
   "ok": true,
   "answer": "the article does not answer how many other readers have asked this question § {#for-the-technically-curious}.",
   "served": "the article does not answer how many other readers have asked this question § {#for-the-technically-curious}.",
   "seconds": 5.5228522180000255,
   "ttft_ms": 5233,
   "tokens_in": 1890,
   "tokens_out": 26,
   "cached_tokens": 32,
   "cites": [
    "for-the-technically-curious"
   ],
   "gates": {
    "cites": [
     "for-the-technically-curious"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "off-17",
   "kind": "off-page"
  },
  {
   "slug": "how-the-beat-lab-works",
   "q": "what phone number reaches support?",
   "ok": true,
   "answer": "the article does not answer the question about a support phone number § {#what-it-is}.",
   "served": "the article does not answer the question about a support phone number § {#what-it-is}.",
   "seconds": 3.264163660999884,
   "ttft_ms": 2807,
   "tokens_in": 30912,
   "tokens_out": 21,
   "cached_tokens": 32,
   "cites": [
    "what-it-is"
   ],
   "gates": {
    "cites": [
     "what-it-is"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "off-18",
   "kind": "off-page"
  },
  {
   "slug": "three-at-the-table",
   "q": "what would this measurement look like at twice the batch size?",
   "ok": true,
   "answer": "the article does not answer this question, though the section § {#the-part-nobody-warns-you-about-the-whole-payroll-needs-an} comes nearest by discussing how memory and speed are priced.",
   "served": "the article does not answer this question, though the section § {#the-part-nobody-warns-you-about-the-whole-payroll-needs-an} comes nearest by discussing how memory and speed are priced.",
   "seconds": 3.3294004220001625,
   "ttft_ms": 565,
   "tokens_in": 7391,
   "tokens_out": 48,
   "cached_tokens": 32,
   "cites": [
    "the-part-nobody-warns-you-about-the-whole-payroll-needs-an"
   ],
   "gates": {
    "cites": [
     "the-part-nobody-warns-you-about-the-whole-payroll-needs-an"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "off-19",
   "kind": "off-page"
  },
  {
   "slug": "the-same-sixteen",
   "q": "which of these models is the best one to use for my own project?",
   "ok": true,
   "answer": "the article does not answer which model is best to use for a project § {#a-different-kind-of-row}. it notes that the question of what to buy is not answered by the reading § {#a-different-kind-of-row}.",
   "served": "the article does not answer which model is best to use for a project § {#a-different-kind-of-row}. it notes that the question of what to buy is not answered by the reading § {#a-different-kind-of-row}.",
   "seconds": 3.2719259440000314,
   "ttft_ms": 721,
   "tokens_in": 6225,
   "tokens_out": 54,
   "cached_tokens": 32,
   "cites": [
    "a-different-kind-of-row"
   ],
   "gates": {
    "cites": [
     "a-different-kind-of-row"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "off-20",
   "kind": "off-page"
  },
  {
   "slug": "seat-trials",
   "q": "how many models will sit this exam next year?",
   "ok": true,
   "answer": "the article does not state how many models will sit this exam next year § {#the-rest-of-the-seminar}.",
   "served": "the article does not state how many models will sit this exam next year § {#the-rest-of-the-seminar}.",
   "seconds": 3.0284716720000233,
   "ttft_ms": 883,
   "tokens_in": 6857,
   "tokens_out": 28,
   "cached_tokens": 32,
   "cites": [
    "the-rest-of-the-seminar"
   ],
   "gates": {
    "cites": [
     "the-rest-of-the-seminar"
    ],
    "redacted": false,
    "figure_unverified": false,
    "uncited": false,
    "new_noun": false,
    "blocked": false,
    "directive_hit": false
   },
   "figures": [],
   "figures_verbatim": true,
   "cites_valid": true,
   "redaction_hits": 0,
   "id": "off-21",
   "kind": "off-page"
  }
 ],
 "verdicts": {
  "on_page_answered": false,
  "non_verbatim_figures": true,
  "off_page_abstained": false,
  "redaction_hits": true,
  "p50_warm_seconds": false,
  "p95_warm_seconds": false
 },
 "passed": false
}