Obsidian vs Logseq — Local-First PKM for Developers (2026)

Pick Obsidian. It has 4,400+ plugins to Logseq’s 555, working mobile apps on iOS and Android, and a stable architecture. Logseq is worth it only if open source is a hard requirement or you think exclusively in outlines and daily journals — and even then, you should know it is mid-transition to an incompatible new database architecture.

Who this is for

Developers choosing a local-first personal knowledge management tool in 2026. If your main concern is team collaboration or real-time editing, neither tool is the right question — see the Notion vs Obsidian comparison instead.

What we tested

Obsidian v1.13.0 (released May 28, 2026; Catalyst Early Access — latest public release is v1.12.1 desktop, v1.12.7 mobile) with community plugin directory at ~4,400 entries. iOS and Android apps on the same v1.13.0 release. Logseq OG v0.10.15 (released December 1, 2025) — the version users download today, with no feature releases since then, only security fixes. Logseq DB — a separate build in beta (desktop; mobile DB is alpha), SQLite-backed and architecturally distinct from OG, in active development.

Obsidian vs Logseq at a glance

	Obsidian	Logseq OG	Logseq DB
Status	Stable, active	No new features since Dec 2025	Beta (desktop) / Alpha (mobile)
Data format	`.md` files	Markdown + EDN	SQLite
Open source	No (free binary)	Yes (AGPLv3)	Yes
Plugin count	~4,400	~555	OG plugins broken
Sync cost	$4/mo annual	~$5/mo (beta)	Not available
Mobile	iOS + Android, v1.13.0	GitHub APK beta only; app stores outdated	iOS alpha only
Local encryption	No — plaintext files	No	No

Core model: notes vs blocks

This is the most important decision. Getting it wrong means fighting the tool every day.

Obsidian is note-first. The note is the atomic unit. Every note is a .md file on your disk with a filename. Links are [[WikiLinks]] between files. The graph view visualizes connections between notes. If you think in documents — one file per concept, one file per meeting, one file per project — Obsidian matches that model directly. Your vault is a folder of files you can open in any editor, grep with ripgrep, or version with Git.

Logseq is block-first. Every bullet point is an independent node with a unique identifier. You link to specific blocks, not just pages. The daily journal is the structural entry point: you open today’s page, write bullets, and the app retrospectively pulls tagged bullets into “pages” that are filtered views of all blocks with a given tag. If you think in continuous streams — daily notes, quick capture, one long outline you reorganize with indentation and tags — Logseq fits that mental model.

Neither model is superior in the abstract. Logseq power users who try Obsidian often complain that separate files feel scattered, that finding yesterday’s half-formed thought requires knowing which note it was in. Obsidian power users who try Logseq often complain that the journal-first structure forces every thought through a date lens, and that pulling canonical knowledge out into clean named documents feels awkward.

A concrete test: how do you capture a meeting note? In Obsidian, you create 2026-06-05 — Arch Review.md and write there. That file exists forever as a named artifact. In Logseq, you open today’s journal, create a block for the meeting, and write nested bullets under it. Both store the same content. Obsidian’s version is easier to find by name, share as a standalone file, or grep. Logseq’s version is easier to query retrospectively — “show me all blocks tagged #arch-review across all journals.”

For developers who work across many projects and need notes that stand alone — code snippets, architectural decisions, API reference scraps — Obsidian’s file-per-note structure composes better with the rest of the toolchain.

Plugin ecosystem and scripting

The gap here is significant: ~4,400 Obsidian plugins vs ~555 Logseq plugins as of June 2026 (Logseq marketplace). That gap isn’t just a count — it reflects six years of Obsidian accumulating plugins for every conceivable workflow while Logseq’s ecosystem never reached the same density.

The highest-leverage Obsidian plugins for developers:

Dataview queries your vault’s YAML frontmatter like a database. You write:

TABLE file.ctime AS Created, status FROM #project WHERE status = "active" SORT file.ctime DESC

Every note with #project and a status: field shows up in the output. Logseq has a built-in query language based on Datalog — more theoretically powerful, considerably harder to learn, and much less documented than Dataview.

Templater runs JavaScript when a new note is created. Pull from an API, format dates, generate YAML frontmatter automatically, call external scripts. The difference from Obsidian’s built-in Templates is that Templater is actually programmable.

obsidian-git commits and pushes your vault on a schedule, without manual action. Vault in GitHub, free cross-device sync, full history. The cost of Obsidian Sync minus the subscription fee. Developers who want to publish from their vault can pair this with a static blog stack that reads markdown directly.

Excalidraw embeds a whiteboard inside any note, stored as a .md file on your disk. Diagrams travel with your vault.

Logseq has plugins that cover some of these — journaling tools, flashcard systems, custom queries. But coverage is thinner, and the deeper issue is architectural: Logseq OG plugins are incompatible with Logseq DB. Every plugin built against the OG file format will need rewriting or replacement when DB reaches general availability. That’s not a migration path — it’s a complete ecosystem reset. If you build automation on Logseq OG today, budget time to rebuild it when DB arrives.

Obsidian plugins run as standard JavaScript in Electron — no sandboxing restrictions beyond what Obsidian’s API exposes. Logseq plugins run in iframes, which improves isolation at the cost of performance and cross-plugin communication.

Mobile and sync costs

Obsidian mobile is the full desktop app ported to iOS and Android, updated to v1.13.0 alongside the desktop release. Plugins work (a subset of them — plugin authors flag mobile support explicitly). The sync options:

Obsidian Sync: $4/mo annual ($48/yr) or $5/mo monthly. End-to-end encrypted, 1 GB storage, version history. Works across every device, no setup beyond entering credentials.
obsidian-git on iOS: Requires Working Copy ($19.99 one-time purchase, or now free for basic use) and some configuration. Free ongoing cost, but iOS doesn’t run background processes, so you push and pull manually or on app open.
iCloud or Dropbox: Works on iOS. Unreliable on Android without third-party apps.

Logseq OG mobile: The App Store and Play Store builds lag the desktop significantly. The latest v0.10.15 (December 2025) ships as a GitHub Android APK beta — outside the official app stores, not as a standard update. The Logseq sync backend (~$5/mo via Open Collective donation) is the only sync path — it’s in beta with no SLA, and the donation model is not a subscription. On Android, there is no iCloud path; Git sync is possible but requires external apps and manual intervention.

Logseq DB mobile: iOS alpha only. Android release has no announced date. Alpha means breaking changes on any release.

If you capture notes on a phone, sync between a laptop and a tablet, or read your vault on mobile — Obsidian wins and it is not close. Obsidian Sync at $48/yr is the lowest-friction path. The app actually works.

One important caveat on the Obsidian side: local vaults have no encryption. Notes are plaintext .md files on your filesystem. Anyone with access to your device or a mounted drive can read them. Obsidian Sync adds E2E encryption in transit and at rest on the sync service, but the local copy is always plaintext. If you store sensitive content — credentials in a notes file, private journal entries, client-confidential notes — this is worth knowing before choosing where to save the vault.

Recommendation matrix

Pick Obsidian if:

You think in named documents, not daily journals — files as the atomic unit maps to how you already work
You rely on or want to build automation (Dataview queries, Templater scripts, obsidian-git pipelines)
Mobile is part of your workflow and needs to actually work today
You want the broadest plugin coverage and the most documentation online
Stability and long-term investment in your tooling matters — Obsidian’s architecture is not in transition

Pick Logseq if:

Open source is a hard requirement and a free binary is a dealbreaker (Obsidian is closed-source freeware)
You think in outlines and daily journals — blocks and the journal-first model match how your brain works
You do not depend on mobile
You are comfortable building on OG knowing migration to DB is coming at an unknown date, or you want to get early on DB despite the beta/alpha instability
You care about graph traversal and backlink queries more than file-level structure

The 2026 wrinkle: users downloading Logseq today get OG v0.10.15 — no feature releases since December 2025, security fixes only. The “real” Logseq (DB) is in beta on desktop and alpha on mobile, with broken plugin compatibility and no Android release. If you’re evaluating Logseq as a platform for the next two to three years, you’re betting on a product that doesn’t exist in a stable form yet.

Caveats

Logseq DB timeline is unknown. The desktop beta has been running for over a year with no general-availability date announced. Logseq OG has had no feature releases since v0.10.15 (December 2025) — security fixes only. There is no stable, documented migration path from OG to DB. Any workflow built on OG today requires a future migration; any workflow built on DB today is subject to breaking changes.

Logseq Sync (~$5/mo) is a donation via Open Collective, not a standard subscription. No SLA, no refund policy, no uptime guarantee. Treat it as a best-effort beta.

obsidian-git on iOS requires Working Copy and manual trigger for push/pull. It is not automatic in the way obsidian-git is on desktop.

Neither Obsidian nor Logseq has an affiliate relationship with toolchew.