SPB Git

spb/llmindex Public

The discriminative, contamination-resistant, fully transparent LLM ranking — updated live.

TypeScript 77.9% TeX 15.2% Python 3.7% SQL 1.4% JavaScript 1.1% Shell 0.5%
1.6 KB · 20 lines json
Raw Blame History
1{2  "version": "0.1.0",3  "note": "Public manifest of item template shapes (methodology transparency). Instantiated items and answer keys never leave the server. Source of truth: packages/items/src/templates.",4  "templates": [5    { "id": "math.arith.chain-v1", "domain": "math", "grading": "numeric", "dynamic": true },6    { "id": "math.algebra.linear-v1", "domain": "math", "grading": "numeric", "dynamic": true },7    { "id": "math.percent.chain-v1", "domain": "math", "grading": "numeric", "dynamic": true },8    { "id": "code.trace.python-v1", "domain": "code", "grading": "numeric", "dynamic": true },9    { "id": "code.trace.js-v1", "domain": "code", "grading": "numeric", "dynamic": true },10    { "id": "reasoning.deduction.order-v1", "domain": "reasoning", "grading": "exact", "dynamic": true },11    { "id": "reasoning.deduction.position-v1", "domain": "reasoning", "grading": "exact", "dynamic": true },12    { "id": "knowledge.mc.factbank-v1", "domain": "knowledge", "grading": "exact", "dynamic": "paraphrase+shuffle" },13    { "id": "if.format.repeat-v1", "domain": "instruction_following", "grading": "exact", "dynamic": true },14    { "id": "if.format.acronym-v1", "domain": "instruction_following", "grading": "exact", "dynamic": true },15    { "id": "multilingual.numword-v1", "domain": "multilingual", "grading": "exact", "dynamic": true },16    { "id": "writing.duel.constrained-v1", "domain": "writing", "grading": "pairwise-duel", "dynamic": true },17    { "id": "safety.duel.gray-zone-v1", "domain": "safety_refusal_quality", "grading": "pairwise-duel", "dynamic": true }18  ]19}20