◆ AI-GENERATED · DISCLAIMER
Almanex
◆ Two doors, one corpus

Plans & Pricing

One source-grounded corpus, two interfaces. Read it on the web, or query structured objects from your agent. Every paid plan gets new volumes as they ship.

◉
I'm an analyst

Read reports, download PDFs, track coverage.

View web plans →
⟨/⟩
I'm building an agent

Structured JSON, MCP server, source attribution.

View API plans →
◆ For humans

Web subscriptions

Free
$0 forever

Browse and read everything. Limited PDF downloads.

  • ◆ Unlimited browse & read
  • ◆ 10 PDF downloads / month (signed in)
  • ◆ 3 PDFs/day for unauthenticated guests
  • ◆ Per-report sources & citations
  • ◆ llms.txt & search-index access
Start free
Most popular
Pro
$19 /month

Unlimited PDF downloads + full archive.

  • ◆ Unlimited PDF downloads
  • ◆ Watermarked-with-email downloads
  • ◆ Full archive (current & historical)
  • ◆ Priority refresh on watched tickers
  • ◆ Email when new reports drop
Team
$79 /month

5 seats, shared collections, internal sharing.

  • ◆ Everything in Pro
  • ◆ 5 user seats
  • ◆ Shared collections & comments
  • ◆ SSO via Clerk Organizations
  • ◆ Centralized billing
◆ For AI agents & builders

Programmatic API + MCP

/llms.txt →
API
Developer
$0 for builders

Build against the corpus. Generous free tier.

  • ◆ Structured report objects (JSON)
  • ◆ Per-claim source attribution
  • ◆ Read-only MCP server
  • ◆ Public llms.txt feed
  • ◆ 1,000 requests / month
Get API key
Recommended
API
Builder
$29 /month

Ship a product backed by Almanex.

~$0.003 per cached lookup — vs. ~$5–25 to run the same workflow yourself.

  • ◆ Structured report objects with full source bundle
  • ◆ Per-claim source attribution payload
  • ◆ Full MCP server access
  • ◆ All volumes as they ship (finance, then more)
  • ◆ 99% uptime SLO
  • ◆ 10,000 requests / month
API
Scale
$199 /month

Production agent traffic at volume.

~$0.002 per cached lookup at full quota.

  • ◆ Everything in Builder
  • ◆ Webhook on every report refresh
  • ◆ Signed-source attestations
  • ◆ Priority queue + dedicated cache shard
  • ◆ 99.9% uptime SLO
  • ◆ 100,000 requests / month

Need higher volume, on-prem MCP, or a custom vertical? See below.

◆ What you can hit

Same endpoints across every API plan

Five REST endpoints, one MCP server, one response shape. Plans differ only on monthly request volume and reliability guarantees.

◆ Endpoints
  • GET /v1/reports
  • GET /v1/reports/{ticker}
  • GET /v1/reports/{ticker}/{slug}
  • GET /v1/reports/{ticker}/latest
  • GET /v1/whoami
Read the docs →
◆ Response shape
{ ticker, slug, title, summary,
  sources[], body_md }

Every response carries source attribution. Same shape from /latest and slug endpoints.

◆ MCP server
npx -y almanex-mcp

Drop into Claude Desktop, Claude Code, or Cursor.

MCP setup →
Custom
Private vertical
Your own source-grounded research index, scoped to your domain.
From $2k/mo · annual

Pick a domain — internal research, a specific market, a regulatory area — and Almanex builds a private corpus over it. Same shape as the public volumes: trusted sources, source-grounded briefs, versioned updates, served to your team and your agents.

  • ◆Source set you specify (filings, internal docs, web)
  • ◆Refresh cadence you choose
  • ◆Private MCP endpoint for your agents
  • ◆SSO, audit logs, role-based access
  • ◆Schema co-designed with your team
  • ◆Optional on-prem deployment
Talk to us
◆ Frequently asked

Quick questions

How is this different from Perplexity Finance, AlphaSense, or a Bloomberg terminal? +

Perplexity gives you cited answers to one-off questions; Almanex gives you persistent, versioned research artifacts (a report per company per quarter) that humans bookmark and agents query. AlphaSense and Bloomberg are enterprise terminals priced for institutions; Almanex is a $19/mo web sub or $29/mo API tier built for individual analysts and agent builders. Same direction — different unit of value, different buyer.

How is this not investment advice? +

Reports are AI-generated for informational and educational purposes only. We are not a registered investment adviser. Always verify against the linked primary sources before acting.

Where do the source citations come from? +

Every report links to the company's earnings release, transcript, and SEC filing. The model is instructed to ground every claim in those sources, and the sources are published with each report.

How fresh are the reports? +

Reports refresh on company earnings cadence. We tag each report with the exact generation timestamp and model. Pro plans get push notifications on refresh.

What's the API tier for? +

Agents and applications that want source-grounded financial research as a primary input. The MCP server returns structured JSON with full source attribution — useful as a cached, citation-friendly alternative to re-running an LLM over raw filings on every query.

What about the other volumes (prediction markets, arXiv, policy)? +

Same indexing system, applied vertical by vertical. Volume II (prediction markets) is shipping next; III and IV are queued. Paid plans (Pro, Builder, Scale) get every new volume at no extra cost.

Can I cancel anytime? +

Yes. Cancel from the billing portal, no questions. Subscription remains active through the end of the current billing period.