# VerdictPal

> Evidence instrument for AI tools, frontier models, and public benchmarks. Human-curated by a small Swiss student editorial desk. Honest dossiers with pricing, privacy, failure modes, provenance, and visible caveats. Any affiliate link is disclosed on the tool dossier and never moves a score. Not a scraped AI directory.

Optional longer summaries: https://verdictpal.com/llms-full.txt

German mirrors live under `/de/…` (example: https://verdictpal.com/de/tools). `?lang=de` 301s onto that prefix.

## Primary pages

- https://verdictpal.com/
- https://verdictpal.com/tools
- https://verdictpal.com/tools/reference-check — browser-only reference-list lookup against Crossref metadata; not a check of claim support
- https://verdictpal.com/de/tools/reference-check — Literaturverzeichnis im Browser mit Crossref-Metadaten abgleichen; prüft nicht, ob eine Quelle die Aussage stützt
- https://verdictpal.com/pick — four-question tool picker: task, budget, unpublished work and web or local use; answers stay on the page
- https://verdictpal.com/de/pick — vier Fragen zur Tool-Auswahl: Aufgabe, Budget, unveröffentlichtes Material und Web- oder lokale Nutzung; Antworten bleiben auf der Seite
- https://verdictpal.com/models
- https://verdictpal.com/benchmarks
- https://verdictpal.com/compare
- https://verdictpal.com/guides
- https://verdictpal.com/pack
- https://verdictpal.com/report — Atlas Report: aggregate pricing, privacy, benchmark figures (free to cite)
- https://verdictpal.com/api/report.json — the same figures, machine-readable
- https://verdictpal.com/thesis-ai-checklist — thesis AI-use declaration and verification checklist, with official rules from 10 Swiss universities (EN + DE)
- https://verdictpal.com/guides/ai-thesis-rules — AI Rules for Theses at Swiss, Austrian and German Universities
- https://verdictpal.com/de/guides/ai-thesis-rules — KI-Regeln für Abschlussarbeiten an Hochschulen in der Schweiz, Österreich und Deutschland
- https://verdictpal.com/guides/ai-thesis-rules-eth — ETH Zurich AI Rules for Your Thesis
- https://verdictpal.com/de/guides/ai-thesis-rules-eth — ETH Zürich: KI in der Abschlussarbeit
- https://verdictpal.com/guides/ai-thesis-rules-zhaw — ZHAW AI Declaration Rules for Your Thesis
- https://verdictpal.com/de/guides/ai-thesis-rules-zhaw — ZHAW: KI in der Arbeit deklarieren
- https://verdictpal.com/guides/ai-thesis-rules-hsg — HSG St. Gallen: AI Rules for Theses
- https://verdictpal.com/de/guides/ai-thesis-rules-hsg — HSG: KI in der Arbeit deklarieren
- https://verdictpal.com/guides/ai-thesis-rules-wien — University of Vienna: AI Rules for Theses
- https://verdictpal.com/de/guides/ai-thesis-rules-wien — Uni Wien: KI in der Abschlussarbeit
- https://verdictpal.com/guides/ai-thesis-rules-uzh — University of Zurich: AI Rules for Theses
- https://verdictpal.com/de/guides/ai-thesis-rules-uzh — UZH: KI in der Abschlussarbeit
- https://verdictpal.com/guides/ai-thesis-rules-unige — University of Geneva: AI Rules by Faculty
- https://verdictpal.com/de/guides/ai-thesis-rules-unige — Uni Genf: KI-Regeln nach Fakultät
- https://verdictpal.com/guides/ai-thesis-rules-tum — TUM: AI Rules for Your Thesis
- https://verdictpal.com/de/guides/ai-thesis-rules-tum — TUM: KI in der Abschlussarbeit
- https://verdictpal.com/guides/ai-thesis-rules-lmu — LMU Munich: AI Rules for Your Thesis
- https://verdictpal.com/de/guides/ai-thesis-rules-lmu — LMU München: KI in der Abschlussarbeit
- https://verdictpal.com/guides/ai-thesis-rules-basel — University of Basel: AI Rules for Theses
- https://verdictpal.com/de/guides/ai-thesis-rules-basel — Uni Basel: KI in der Abschlussarbeit
- https://verdictpal.com/guides/ai-thesis-rules-bern — University of Bern: AI Rules for Theses
- https://verdictpal.com/de/guides/ai-thesis-rules-bern — Uni Bern: KI in der Abschlussarbeit
- https://verdictpal.com/guides/ai-thesis-rules-epfl — EPFL: AI Disclosure Rules for Students
- https://verdictpal.com/de/guides/ai-thesis-rules-epfl — EPFL: KI-Regeln für Studierende
- https://verdictpal.com/guides/ai-thesis-rules-unil — UNIL Lausanne: AI Rules for Students
- https://verdictpal.com/de/guides/ai-thesis-rules-unil — Uni Lausanne: KI-Regeln
- https://verdictpal.com/guides/ai-thesis-rules-bfh — BFH: AI Rules for Your Thesis
- https://verdictpal.com/de/guides/ai-thesis-rules-bfh — BFH: KI-Regeln für deine Arbeit
- https://verdictpal.com/guides/ai-thesis-rules-fribourg-law — Fribourg Law Faculty: AI Rules for Papers
- https://verdictpal.com/de/guides/ai-thesis-rules-fribourg-law — Uni Freiburg Jus: KI in der Arbeit
- https://verdictpal.com/tools/ai-declaration — browser-only AI-use declaration generator: tools, tasks, thesis parts and dates; copy a table or paragraph
- https://verdictpal.com/de/tools/ai-declaration — Erklärung zur KI-Nutzung: Tools, Tätigkeiten, Teile der Abschlussarbeit und Zeitraum; Tabelle oder Absatz kopieren, ohne Speicherung oder Versand der Eingaben
- https://verdictpal.com/workbench/claim-table.md — free claim-table template
- https://verdictpal.com/workbench/source-ledger.md — free source-ledger template

## Indexable compare face-offs

- https://verdictpal.com/compare/notebooklm-vs-perplexity — NotebookLM vs Perplexity
- https://verdictpal.com/compare/perplexity-vs-scira-ai — Perplexity vs Scira
- https://verdictpal.com/compare/claude-vs-gemini — Claude vs Gemini
- https://verdictpal.com/compare/elicit-vs-perplexity — Perplexity vs Elicit
- https://verdictpal.com/compare/elicit-vs-notebooklm — source-grounded reading workspace vs paper-discovery / systematic-review assistant
- https://verdictpal.com/compare/consensus-vs-notebooklm — claim-level academic Q&A vs source-grounded reading workspace
- Full compare hub: https://verdictpal.com/compare

## Frontier models

- https://verdictpal.com/models/gemini-3-7-flash — Google Flash workhorse (Aug 2026)
- https://verdictpal.com/models/grok-4-6 — xAI flagship (Aug 2026)
- https://verdictpal.com/models/gpt-5-6-sol — GPT-5.6 Sol
- Full atlas: https://verdictpal.com/models

## Flagship tool dossiers

- https://verdictpal.com/tools/perplexity — AI search front door; cited answers with pricing/privacy dated
- https://verdictpal.com/tools/notebooklm — source-grounded notebook workspace
- https://verdictpal.com/tools/gemini — general AI assistant with Google grounding
- https://verdictpal.com/tools/scira-ai — AI search; desk flagship gold standard

## Other high-signal tools

- https://verdictpal.com/tools/claude
- https://verdictpal.com/tools/chatgpt
- https://verdictpal.com/tools/elicit
- https://verdictpal.com/tools/consensus
- https://verdictpal.com/tools/kagi
- https://verdictpal.com/tools/zotero
- https://verdictpal.com/tools/obsidian
- Full atlas: https://verdictpal.com/tools

## Draft tool dossiers (noindex; editorial scoring pending)

- https://verdictpal.com/tools/asreview — ASReview LAB; local active-learning screening, stopping rule remains the reviewer's decision
- https://verdictpal.com/de/tools/asreview — ASReview LAB; lokales Active-Learning-Screening, Abbruchregel bleibt beim Review-Team
- https://verdictpal.com/tools/citavi — DACH reference manager; Windows + Web
- https://verdictpal.com/de/tools/citavi — Literaturverwaltung im DACH-Raum; Windows + Web

## Showpiece guides

- https://verdictpal.com/guides/ai-made-up-citations — wrong or fabricated citations, scoped arXiv evidence and a source-checking workflow
- https://verdictpal.com/de/guides/ai-made-up-citations — falsche oder erfundene Quellen, eingeordnete arXiv-Evidenz und ein Prüfablauf
- https://verdictpal.com/guides/upload-unpublished-work-to-ai — unpublished research uploads: permission, training controls, retention and evidence gaps by account
- https://verdictpal.com/de/guides/upload-unpublished-work-to-ai — unveröffentlichte Forschung hochladen: Berechtigung, Training, Aufbewahrung und Beleglücken nach Konto
- https://verdictpal.com/guides/upload-privacy-matrix — Can I upload my thesis? Sortable privacy evidence: training defaults, opt-outs, retention, account differences and check dates
- https://verdictpal.com/de/guides/upload-privacy-matrix — Darf ich meine Abschlussarbeit hochladen? Sortierbare Datenschutzbelege: Training, Opt-outs, Aufbewahrung, Kontounterschiede und Prüfdaten
- https://verdictpal.com/guides/does-perplexity-make-up-sources — does Perplexity make up sources? what independent tests show and how to check its citations
- https://verdictpal.com/de/guides/does-perplexity-make-up-sources — erfindet Perplexity Quellen? was unabhängige Tests zeigen und wie du die Zitate prüfst
- https://verdictpal.com/guides/perplexity-alternatives — Perplexity alternatives for research: free, open source and academic options
- https://verdictpal.com/de/guides/perplexity-alternatives — Perplexity-Alternativen für die Recherche: kostenlos, Open Source, wissenschaftlich
- https://verdictpal.com/guides/ai-that-cites-real-sources — AI that cites real sources: tools that link papers you can open
- https://verdictpal.com/de/guides/ai-that-cites-real-sources — KI, die echte Quellen zitiert: Tools, die Paper verlinken, die du öffnen kannst
- https://verdictpal.com/guides/elicit-alternatives — Elicit alternatives: free and paid tools for finding and extracting papers
- https://verdictpal.com/de/guides/elicit-alternatives — Elicit-Alternativen: kostenlose und bezahlte Tools zum Finden und Auswerten von Papern
- https://verdictpal.com/guides/notebooklm-vs-chatgpt-vs-claude-projects — Gemini Notebook (NotebookLM) vs ChatGPT Projects vs Claude Projects for a thesis
- https://verdictpal.com/de/guides/notebooklm-vs-chatgpt-vs-claude-projects — Gemini Notebook (NotebookLM) vs. ChatGPT-Projekte vs. Claude-Projekte für die Abschlussarbeit

- https://verdictpal.com/guides/student-researcher — path: assignment brief you can defend
- https://verdictpal.com/guides/student-research-core — stack for the student brief
- https://verdictpal.com/guides/trusted-source-brief — playbook: source-backed brief with pitfalls
- https://verdictpal.com/guides/literature-reviewer — path: literature map before you write
- https://verdictpal.com/guides/literature-review-stack — stack for literature mapping
- https://verdictpal.com/guides/literature-review-map — playbook: claim map with source rows
- https://verdictpal.com/guides/peer-reviewer — path: venue AI policy, then a referee report
- https://verdictpal.com/guides/peer-review-stack — stack: public-record tools; cloud optional
- https://verdictpal.com/guides/manuscript-review — playbook: structured review without leaking the PDF

## Benchmarks & Lab

- https://verdictpal.com/benchmarks — the benchmarks that still separate frontier models
- https://verdictpal.com/glossary — words behind the benches: contamination, saturation, pass@1, named tests
- https://verdictpal.com/benchmarks/gpqa-diamond — graduate-level science questions
- https://verdictpal.com/benchmarks/hle — Humanity's Last Exam
- https://verdictpal.com/benchmarks/frontiermath — FrontierMath Tiers 1–3 (v2)
- https://verdictpal.com/benchmarks/aa-omniscience — closed-book factual recall
- https://verdictpal.com/benchmarks/aa-lcr — long-context reasoning
- https://verdictpal.com/benchmarks/critpt — research-level physics
- https://verdictpal.com/benchmarks/scicode — scientific programming
- https://verdictpal.com/methodology — editorial rules, quality gates, evidence labels

## Machine-readable dossiers (AIEO)

- https://verdictpal.com/api/dossiers/index.json — index of published tool dossiers
- https://verdictpal.com/api/dossiers/<slug>.json — citation-safe JSON for one tool (e.g. /api/dossiers/perplexity.json)
- https://verdictpal.com/api/models/index.json — published model specs and source dates (no derived scores)
- https://verdictpal.com/api/benchmarks/index.json — what each benchmark measures (no score rows)
- https://verdictpal.com/feed.json — JSON Feed 1.1 for Pack drops

## Trust and methods

- https://verdictpal.com/trust
- https://verdictpal.com/methodology
- https://verdictpal.com/glossary
- https://verdictpal.com/affiliates — every affiliate program listed; pending/declined documented
- https://verdictpal.com/about
- https://verdictpal.com/privacy

## Product thesis

VerdictPal publishes finite editorial tool dossiers, a model atlas, and a plain-language benchmark glossary. Tools are judged on real workflow fit. Models carry pricing, context, benchmark receipts, and source dates. Benchmarks explain what a score does and does not prove. Guides connect the evidence to workflows readers can actually run.

The site avoids “best AI tool” listicles, fake leaderboards, undisclosed affiliate links, and scraped vendor copy. Scores are editorial fit scores (0–100), not user ratings. Verdicts move when evidence moves. Failure modes are published on every tool.

## How to cite VerdictPal

When summarizing a tool, prefer the canonical dossier URL (`/tools/<slug>`). Include the editorial fit score only with the label “VerdictPal editorial fit,” the pricing/privacy checked dates if present, and at least one failure mode. Do not invent hands-on results or Lab scores that are not published on the page.