Tardis-style historical data infrastructure and 0xArchive overlap in replay and historical-market-data jobs, but differ by venue focus, route model, and supported workflow boundaries. The choice is usually breadth versus supported-venue implementation speed and route specificity.Documentation Index
Fetch the complete documentation index at: https://docs.0xarchive.io/llms.txt
Use this file to discover all available pages before exploring further.
Comparison
| Need | Tardis-style broad vendor | 0xArchive |
|---|---|---|
| Broad exchange history | Strong fit | Not the primary product job |
| Download-first historical workflows | Strong fit | Use Data Catalog when supported |
| Hyperliquid core route clarity | Varies by vendor abstraction | First-class route family |
| Hyperliquid Spot, HIP-3, HIP-4 | Depends on vendor coverage | First-class Hyperliquid-scoped families |
| Lighter L2/L3 | Depends on vendor coverage | First-class Lighter route family |
| Coding-agent route safety | Depends on docs and spec | OpenAPI, Markdown, llms.txt, CLI, SDK, MCP, Skill |
When 0xArchive Wins
Choose 0xArchive when the workflow is specifically Hyperliquid core, Hyperliquid Spot, HIP-3, HIP-4, or Lighter and the team values exact routes, examples, freshness gates, replay/reconstruction guidance, and agent/codegen support.When Tardis May Win
Choose a broad historical data vendor when cross-exchange breadth is more important than supported-venue depth, or when the procurement requirement is a wide downloadable catalog across many exchanges.Breadth Versus Route Fit
The decision should name the trade directly. A broad vendor can reduce vendor count when the team needs many exchanges in one purchase. 0xArchive can reduce implementation work when the team needs a smaller supported venue set with route-specific docs, agent context, replay guidance, and data-quality checks attached to the same workflow.Evaluation Checklist
Do not compare these options only by whether both can return historical rows. Ask what the team must ship after the purchase. If the deliverable is a cross-exchange research lake, a broad vendor can be the simpler buying answer. If the deliverable is a product, agent, backtest, or internal workflow around Hyperliquid core, Spot, HIP-3, HIP-4, or Lighter, route semantics become the deciding factor. For 0xArchive, evaluate whether the required market family has the exact REST route, WebSocket channel, export schema, freshness signal, and example needed by the implementation. For a broad vendor, evaluate whether the vendor exposes the same supported venue family with enough depth and symbol semantics, or whether the buyer will still need a translation layer before the data is usable.Implementation Notes
The highest-risk migration mistake is treating vendor-normalized symbols as interchangeable with 0xArchive venue-family routes. Preserve the source family in storage. A Hyperliquid core perp, a Spot pair, a HIP-3 builder market, a HIP-4 outcome market, and a Lighter market should not share a genericexchange_symbol field without another column that records the original family and route.
If the evaluation request includes replay or reconstruction, test sequence metadata before testing volume. A large historical download that cannot reproduce book state is not a replacement for a replay-ready workflow.