HLL 0.1-TX
Grades, item by item
Every graded answer, every axis, every citation check. Scores are mechanical and recomputable. Public-split items show the blinded facts and the model's full answer; private-split items show scores and notes only. Weights: Standard 2, Authority 2, Outcome 2, Quote fidelity 2. A fabricated citation caps the item at 25%.
watching: refreshes every 30 s|all runsdeepseek/deepseek-v4-flash · rawdeepseek/deepseek-v4-flash · juiceddeepseek/deepseek-v4-flash · dr2upstage/solar-pro4 · rawupstage/solar-pro4 · dr2openai/gpt-oss-120b · rawopenai/gpt-oss-20b · rawopenai/gpt-oss-120b · dr2openai/gpt-oss-20b · dr2google/gemini-3.7-flash · rawgoogle/gemini-3.7-flash · dr2
administrativeappellate-mandamusappellate-procedurearbitrationassault-iiedauto-accidentauto-insurance-uimchild-custodychild-supportcivil-litigationconstitutionalconstructionconsumer-dtpacontractsconversioncps-terminationcriminaldefamationdiscoverydivorce-propertyeasements-boundarieselectionseminent-domainemploymentevictionevidencefamilyforeclosurefraudgovernment-municipalgroundwaterhabeashealthcare-medmalhoa-condohomicideineffective-assistanceinsurancejurisdiction-traplaborlandlord-tenantlegal-malpracticelimitationsnuisance-trespassoil-gasopen-recordsotherpersonal-injurypremises-liabilityprobateproducts-liabilityprofessional-responsibilityreal-estate-brokersreal-propertyschool-lawsearch-seizuresovereign-immunitysummary-judgmenttaxtheft-fraud-crimestorttortious-interferencetruckingvenue-forumworkers-compwrongful-death
50%hll-cl11229890-3appellate-mandamusdifficulty 3private splitgoogle/gemini-3.7-flash · dr2
Axes
Outcome100%×2
Standard-×2
Authority0%×2
Quote fidelity-×2
Citation checks
found827 S.W.2d 833 · in DocketRouter index
Private-split item: facts, answer key and answer withheld so the leaderboard stays uncontaminated. The scores above are complete.
33534ms · graded 2026-08-28 22:28
50%hll-cl11229890-1appellate-mandamusdifficulty 4private splitgoogle/gemini-3.7-flash · dr2
Axes
Outcome100%×2
Standard-×2
Authority0%×2
Quote fidelity-×2
Citation checks
found186 S.W.3d 558 · in DocketRouter index
found940 S.W.2d 86 · in DocketRouter index
Private-split item: facts, answer key and answer withheld so the leaderboard stays uncontaminated. The scores above are complete.
37302ms · graded 2026-08-28 22:28
92%hll-cl11330761-1appellate-mandamusdifficulty 4private splitgoogle/gemini-3.7-flash · raw
Axes
Outcome100%×2
Standard100%×2
Authority67%×2
Quote fidelity100%×2
Citation checks
found233 S.W.3d 835 · in DocketRouter index
in opinion207 S.W.3d 334
found148 S.W.3d 124 · in DocketRouter index
in opinion133 S.W.3d 217
Quote fidelity (every quotation must exist verbatim in its source)
supported233 S.W.3d 835 for: grants or denies a plea to the jurisdiction by a governmental unit.
Private-split item: facts, answer key and answer withheld so the leaderboard stays uncontaminated. The scores above are complete.
15859ms · graded 2026-08-28 21:56
100%hll-cl11229890-3appellate-mandamusdifficulty 3private splitgoogle/gemini-3.7-flash · raw
Axes
Outcome100%×2
Standard-×2
Authority100%×2
Quote fidelity100%×2
Citation checks
found827 S.W.2d 833 · in DocketRouter index
found270 S.W.3d 757 · in DocketRouter index
found335 S.W.3d 808 · in DocketRouter index
Quote fidelity (every quotation must exist verbatim in its source)
supported827 S.W.2d 833 for: [A] certified or sworn copy of every document that is material to the relator's claim for relief and that was filed in any underlying procee
Private-split item: facts, answer key and answer withheld so the leaderboard stays uncontaminated. The scores above are complete.
22317ms · graded 2026-08-28 21:18
50%hll-cl11229890-1appellate-mandamusdifficulty 4private splitgoogle/gemini-3.7-flash · raw
Axes
Outcome100%×2
Standard-×2
Authority0%×2
Quote fidelity-×2
Citation checks
found148 S.W.3d 124 · in DocketRouter index
found164 S.W.3d 379 · in DocketRouter index
found186 S.W.3d 534 · in DocketRouter index
found786 S.W.2d 248 · in DocketRouter index
found111 S.W.3d 56 · in DocketRouter index
Private-split item: facts, answer key and answer withheld so the leaderboard stays uncontaminated. The scores above are complete.
19108ms · graded 2026-08-28 21:18
0%hll-cl11330761-1appellate-mandamusdifficulty 4private splitblankopenai/gpt-oss-20b · dr2
Axes
Outcome0%×2
Standard0%×2
Authority0%×2
Quote fidelity-×2
Citation checks
no citations offered
Private-split item: facts, answer key and answer withheld so the leaderboard stays uncontaminated. The scores above are complete.
54764ms · graded 2026-08-28 19:41
19%hll-cl11330761-1appellate-mandamusdifficulty 4private splitopenai/gpt-oss-120b · dr2
Axes
Outcome0%×2
Standard0%×2
Authority58%×2
Quote fidelity-×2
Citation checks
found233 S.W.3d 835 · in DocketRouter index
Private-split item: facts, answer key and answer withheld so the leaderboard stays uncontaminated. The scores above are complete.
40581ms · graded 2026-08-28 17:31
65%hll-cl11330761-1appellate-mandamusdifficulty 4private splitdeepseek/deepseek-v4-flash · dr2
Axes
Outcome100%×2
Standard0%×2
Authority58%×2
Quote fidelity100%×2
Citation checks
found233 S.W.3d 835 · in DocketRouter index
found827 S.W.2d 833 · in DocketRouter index
Quote fidelity (every quotation must exist verbatim in its source)
supported for: grants or denies a plea to the jurisdiction by a governmental unit.
Private-split item: facts, answer key and answer withheld so the leaderboard stays uncontaminated. The scores above are complete.
29580ms · graded 2026-08-28 15:39
65%hll-cl11330761-1appellate-mandamusdifficulty 4private splitupstage/solar-pro4 · dr2
Axes
Outcome0%×2
Standard100%×2
Authority58%×2
Quote fidelity100%×2
Citation checks
found827 S.W.2d 833 · in DocketRouter index
found34 S.W.3d 547 · in DocketRouter index
found233 S.W.3d 835 · in DocketRouter index
Quote fidelity (every quotation must exist verbatim in its source)
no passage34 S.W.3d 547 for: grants or denies a plea to the jurisdiction by a governmental unit as that term is defined in Section 101.001.
supported for: grants or denies a plea to the jurisdiction by a governmental unit
Private-split item: facts, answer key and answer withheld so the leaderboard stays uncontaminated. The scores above are complete.
67510ms · graded 2026-08-28 15:22
50%hll-cl11229890-3appellate-mandamusdifficulty 3private splitopenai/gpt-oss-20b · dr2
Axes
Outcome100%×2
Standard-×2
Authority0%×2
Quote fidelity-×2
Citation checks
no citations offered
Private-split item: facts, answer key and answer withheld so the leaderboard stays uncontaminated. The scores above are complete.
50414ms · graded 2026-08-28 14:55
25%hll-cl11229890-1appellate-mandamusdifficulty 4private splitopenai/gpt-oss-20b · dr2
Axes
Outcome100%×2
Standard-×2
Authority0%×2
Quote fidelity0%×2
Citation checks
found701 S.W.2d 238 · in DocketRouter index
found876 S.W.2d 304 · in DocketRouter index
Quote fidelity (every quotation must exist verbatim in its source)
unsupported for: in a position of extraordinary disadvantage
Private-split item: facts, answer key and answer withheld so the leaderboard stays uncontaminated. The scores above are complete.
101276ms · graded 2026-08-28 14:54
100%hll-cl11229890-3appellate-mandamusdifficulty 3private splitopenai/gpt-oss-120b · dr2
Axes
Outcome100%×2
Standard-×2
Authority100%×2
Quote fidelity-×2
Citation checks
found141 S.W.3d 144 · in DocketRouter index
found236 S.W.3d 207 · in DocketRouter index
Private-split item: facts, answer key and answer withheld so the leaderboard stays uncontaminated. The scores above are complete.
39733ms · graded 2026-08-28 13:59
50%hll-cl11229890-1appellate-mandamusdifficulty 4private splitopenai/gpt-oss-120b · dr2
Axes
Outcome100%×2
Standard-×2
Authority0%×2
Quote fidelity-×2
Citation checks
found701 S.W.2d 238 · in DocketRouter index
found148 S.W.3d 124 · in DocketRouter index
Private-split item: facts, answer key and answer withheld so the leaderboard stays uncontaminated. The scores above are complete.
36318ms · graded 2026-08-28 13:58
25%hll-cl11229890-3appellate-mandamusdifficulty 3private splitdeepseek/deepseek-v4-flash · dr2
Axes
Outcome100%×2
Standard-×2
Authority0%×2
Quote fidelity0%×2
Citation checks
found827 S.W.2d 833 · in DocketRouter index
Quote fidelity (every quotation must exist verbatim in its source)
unsupported for: a certified or sworn copy of every document that is material to the relator's claim for relief and that was filed in any underlying proceedi
Private-split item: facts, answer key and answer withheld so the leaderboard stays uncontaminated. The scores above are complete.
38052ms · graded 2026-08-28 11:38
50%hll-cl11229890-1appellate-mandamusdifficulty 4private splitdeepseek/deepseek-v4-flash · dr2
Axes
Outcome100%×2
Standard-×2
Authority0%×2
Quote fidelity-×2
Citation checks
found148 S.W.3d 124 · in DocketRouter index
found71 S.W.3d 313 · in DocketRouter index
Private-split item: facts, answer key and answer withheld so the leaderboard stays uncontaminated. The scores above are complete.
102764ms · graded 2026-08-28 11:38
25%hll-cl11229890-3appellate-mandamusdifficulty 3private splitupstage/solar-pro4 · dr2
Axes
Outcome100%×2
Standard-×2
Authority0%×2
Quote fidelity0%×2
Citation checks
found141 S.W.3d 144 · in DocketRouter index
Quote fidelity (every quotation must exist verbatim in its source)
unsupported for: generally fails to comply with Texas Rule of Appellate Procedure 52.3 and fails to include either an appendix or a record
Private-split item: facts, answer key and answer withheld so the leaderboard stays uncontaminated. The scores above are complete.
25518ms · graded 2026-08-28 10:44
25%hll-cl11229890-1appellate-mandamusdifficulty 4private splitupstage/solar-pro4 · dr2
Axes
Outcome100%×2
Standard-×2
Authority0%×2
Quote fidelity0%×2
Citation checks
found701 S.W.2d 238 · in DocketRouter index
found876 S.W.2d 304 · in DocketRouter index
Quote fidelity (every quotation must exist verbatim in its source)
unsupported for: facial constitutional challenge to § 172.021(g) entitles a candidate to mandamus without showing compliance or seeking a cure,
Private-split item: facts, answer key and answer withheld so the leaderboard stays uncontaminated. The scores above are complete.
49896ms · graded 2026-08-28 10:43
0%hll-cl11330761-1appellate-mandamusdifficulty 4private splitblankopenai/gpt-oss-20b · raw
Axes
Outcome0%×2
Standard0%×2
Authority0%×2
Quote fidelity-×2
Citation checks
no citations offered
Private-split item: facts, answer key and answer withheld so the leaderboard stays uncontaminated. The scores above are complete.
56857ms · graded 2026-08-28 10:11
0%hll-cl11229890-1appellate-mandamusdifficulty 4private splitblankopenai/gpt-oss-20b · raw
Axes
Outcome0%×2
Standard-×2
Authority0%×2
Quote fidelity-×2
Citation checks
no citations offered
Private-split item: facts, answer key and answer withheld so the leaderboard stays uncontaminated. The scores above are complete.
43112ms · graded 2026-08-28 09:38
0%hll-cl11229890-3appellate-mandamusdifficulty 3private splitblankopenai/gpt-oss-20b · raw
Axes
Outcome0%×2
Standard-×2
Authority0%×2
Quote fidelity-×2
Citation checks
no citations offered
Private-split item: facts, answer key and answer withheld so the leaderboard stays uncontaminated. The scores above are complete.
10079ms · graded 2026-08-28 09:38
15%hll-cl11330761-1appellate-mandamusdifficulty 4private splitopenai/gpt-oss-120b · raw
Axes
Outcome0%×2
Standard0%×2
Authority58%×2
Quote fidelity0%×2
Citation checks
found361 S.W.3d 1 · in CourtListener bulk table
Quote fidelity (every quotation must exist verbatim in its source)
unsupported for: *clear legal right* to the performance of a ministerial duty
unsupported for: *the appellate jurisdiction conferred by § 51.014(a)(8) is limited to orders that are expressly appealable under § 51.014(a)(5)*.
unsupported for: *the trial court’s proceeding to trial after refusing to rule is a jurisdictional error that must be corrected by mandamus, not by an interl
Private-split item: facts, answer key and answer withheld so the leaderboard stays uncontaminated. The scores above are complete.
82011ms · graded 2026-08-28 09:08
25%hll-cl11229890-1appellate-mandamusdifficulty 4private splitopenai/gpt-oss-120b · raw
Axes
Outcome100%×2
Standard-×2
Authority0%×2
Quote fidelity0%×2
Citation checks
found247 S.W.3d 1 · in CourtListener bulk table
Quote fidelity (every quotation must exist verbatim in its source)
no passage247 S.W.3d 1 for: a petition for mandamus must be filed within a reasonable time after the right to relief accrues
unsupported for: state with particularity the duty to be performed, the right to be enforced, and the facts showing that the duty is clear and that the petit
unsupported for: deadline for filing a candidate’s application
Private-split item: facts, answer key and answer withheld so the leaderboard stays uncontaminated. The scores above are complete.
43769ms · graded 2026-08-28 08:29
25%hll-cl11229890-3appellate-mandamusdifficulty 3private splitopenai/gpt-oss-120b · raw
Axes
Outcome100%×2
Standard-×2
Authority100%×2
Quote fidelity0%×2
Citation checks
no citations offered
Quote fidelity (every quotation must exist verbatim in its source)
unsupported for: The record in a mandamus proceeding shall consist of the original petition, the order or judgment sought to be reviewed, and any other docum
unsupported for: a mandamus petition that fails to attach the complete underlying application and petition is procedurally deficient; the defect is fatal and
unsupported for: the completeness of the record is a jurisdictional prerequisite for the issuance of a writ of mandamus.
unsupported for: Rule 52.7(a) is not a mere evidentiary rule; it is a procedural prerequisite that, if not satisfied, deprives the court of the authority to
unsupported for: a petition that does not comply with the procedural requisites of Rule 52.7(a) may be dismissed *on its own* without addressing the merits.
Private-split item: facts, answer key and answer withheld so the leaderboard stays uncontaminated. The scores above are complete.
24583ms · graded 2026-08-28 08:29
25%hll-cl11330761-1appellate-mandamusdifficulty 4private split · capped: fabricated citationupstage/solar-pro4 · raw
Axes
Outcome0%×2
Standard100%×2
Authority0%×2
Quote fidelity100%×2
Citation checks
found353 S.W.3d 128 · in DocketRouter index
unverified536 S.W.3d 221 · verifier request budget exhausted
unverified499 S.W.3d 594 · verifier request budget exhausted
not found148 S.W.3d 293 · cached; confirmed on retry
Quote fidelity (every quotation must exist verbatim in its source)
supported353 S.W.3d 128 for: denies a plea to the jurisdiction.
Private-split item: facts, answer key and answer withheld so the leaderboard stays uncontaminated. The scores above are complete.
31784ms · graded 2026-08-28 07:52
75%hll-cl11330761-1appellate-mandamusdifficulty 4private splitdeepseek/deepseek-v4-flash · raw
Axes
Outcome100%×2
Standard100%×2
Authority0%×2
Quote fidelity100%×2
Citation checks
found523 S.W.3d 103 · in DocketRouter index
found501 S.W.3d 71 · in DocketRouter index
found579 S.W.3d 431 · in CourtListener bulk table
Quote fidelity (every quotation must exist verbatim in its source)
supported523 S.W.3d 103 for: grants or denies a plea to the jurisdiction by a governmental unit.
Private-split item: facts, answer key and answer withheld so the leaderboard stays uncontaminated. The scores above are complete.
45936ms · graded 2026-08-28 07:25
50%hll-cl11229890-1appellate-mandamusdifficulty 4private splitupstage/solar-pro4 · raw
Axes
Outcome100%×2
Standard-×2
Authority0%×2
Quote fidelity-×2
Citation checks
unverified140 S.W.3d 849 · verifier rate-limited
unverified444 S.W.3d 369 · verifier rate-limited
Private-split item: facts, answer key and answer withheld so the leaderboard stays uncontaminated. The scores above are complete.
77646ms · graded 2026-08-28 07:09
50%hll-cl11229890-3appellate-mandamusdifficulty 3private splitupstage/solar-pro4 · raw
Axes
Outcome100%×2
Standard-×2
Authority0%×2
Quote fidelity-×2
Citation checks
found140 S.W.3d 691 · in DocketRouter index
Private-split item: facts, answer key and answer withheld so the leaderboard stays uncontaminated. The scores above are complete.
53166ms · graded 2026-08-28 07:09
50%hll-cl11229890-1appellate-mandamusdifficulty 4private splitdeepseek/deepseek-v4-flash · raw
Axes
Outcome100%×2
Standard-×2
Authority0%×2
Quote fidelity-×2
Citation checks
found148 S.W.3d 124 · in DocketRouter index
found71 S.W.3d 313 · in DocketRouter index
found576 U.S. 409 · in DocketRouter index
found356 S.W.3d 455 · in DocketRouter index
found146 S.W.3d 648 · in DocketRouter index
found479 U.S. 189 · in DocketRouter index
Private-split item: facts, answer key and answer withheld so the leaderboard stays uncontaminated. The scores above are complete.
89750ms · graded 2026-08-28 06:57
50%hll-cl11229890-3appellate-mandamusdifficulty 3private splitdeepseek/deepseek-v4-flash · raw
Axes
Outcome100%×2
Standard-×2
Authority0%×2
Quote fidelity-×2
Citation checks
found148 S.W.3d 124 · in DocketRouter index
found328 S.W.3d 883 · in DocketRouter index
found270 S.W.3d 757 · in DocketRouter index
found492 S.W.3d 300 · in DocketRouter index
Quote fidelity (every quotation must exist verbatim in its source)
no passage148 S.W.3d 124 for: a certified or sworn copy of every document that is material to the relator’s claim and that was filed in any underlying proceeding.
Private-split item: facts, answer key and answer withheld so the leaderboard stays uncontaminated. The scores above are complete.
44530ms · graded 2026-08-28 06:57
67%hll-flp11166857-1appellate-mandamusdifficulty 4private splitdeepseek/deepseek-v4-flash · dr2
Axes
Outcome100%×2
Standard100%×2
Authority0%×2
Quote fidelity-×2
Citation checks
found827 S.W.2d 833 · in DocketRouter index
found618 S.W.2d 535 · in DocketRouter index
unverified478 S.W.3d 639 · search unavailable
found492 S.W.3d 287 · in DocketRouter index
Private-split item: facts, answer key and answer withheld so the leaderboard stays uncontaminated. The scores above are complete.
47959ms · graded 2026-08-27 21:02
0%hll-flp11062955-1appellate-mandamusdifficulty 4private splitdeepseek/deepseek-v4-flash · dr2
Axes
Outcome0%×2
Standard-×2
Authority0%×2
Quote fidelity-×2
Citation checks
no citations offered
Private-split item: facts, answer key and answer withheld so the leaderboard stays uncontaminated. The scores above are complete.
41030ms · graded 2026-08-27 20:44
67%hll-flp11166857-1appellate-mandamusdifficulty 4private splitdeepseek/deepseek-v4-flash · raw
Axes
Outcome100%×2
Standard100%×2
Authority0%×2
Quote fidelity-×2
Citation checks
found60 S.W.3d 118 · in DocketRouter index
found492 S.W.3d 300 · in DocketRouter index
Private-split item: facts, answer key and answer withheld so the leaderboard stays uncontaminated. The scores above are complete.
30619ms · graded 2026-08-27 18:59
0%hll-flp11062955-1appellate-mandamusdifficulty 4private splitblankdeepseek/deepseek-v4-flash · raw
Axes
Outcome0%×2
Standard-×2
Authority0%×2
Quote fidelity-×2
Citation checks
no citations offered
Private-split item: facts, answer key and answer withheld so the leaderboard stays uncontaminated. The scores above are complete.
30964ms · graded 2026-08-27 16:36