Databases

Where Veil's data actually comes from.

Every literature API and platform integration this workspace uses or plans to use, labeled by what's real today versus what's on the roadmap — not a decorative logo wall.

2 connected · 11 planned · 13 total

Literature

Literature & citation APIs

OpenAlex

Open academic knowledge graph — works, authors, venues.

app/api/papers/route.tsconnected

Tavily

General web search, used for broader literature discovery.

app/api/search/route.tsconnected

Semantic Scholar

Citation graph, influential-citation signal, TL;DR summaries.

planned

Crossref

DOI registry — canonical bibliographic metadata.

planned

arXiv

Preprint search and full text for cs/physics/math/stat.

planned

Unpaywall

Legal open-access PDF resolution for paywalled DOIs.

planned

Europe PMC

Biomedical literature — the domain pack for Biology mode.

planned

Platform

Compute & platform integrations

GitHub

Version control for training runs and generated artifacts.

planned

Hugging Face

Model and dataset hub.

planned

Weights & Biases

Experiment tracking for fine-tuning runs.

planned

Modal

On-demand GPU compute for training/fine-tuning jobs.

planned

Prime Intellect

Distributed GPU compute marketplace.

planned

Pinecone

Managed vector search — an alternative to the self-hosted pgvector retrieval stack.

planned

“Connected” means a route in this codebase calls that API today. “Planned” means it’s named in this project’s own build spec as a target but nothing calls it yet. Nothing on this page is a claim about the future dressed up as the present.