ToolJet
Observability lands on OpenTelemetry semconv in the LTS train
A side-by-side editorial comparison of authentik and missSBM — release velocity, themes, recent moves, and the top alternatives to consider.
authentik 2026.8 ships: Actors, domain-joined Agents, and a push past browser-mediated SSO
2026.8.0 is out, closing a seven-candidate train that ran through early August. The GA tag itself is the last cherry-pick batch — SCIM group membership removals, a proxy redirect that preserves query strings, session deletion on user deactivation — but the release it finalizes is where the substance lives: an Actors primitive in core, an enterprise Agent requiring a domain join and its own API scope, OAuth2 token exchange delegation, and a CAS source integration.
missSBM returns after four dormant years with a stricter API and a new refinement step.
The package fits stochastic block models to networks with missing data, covering both missing-at-random and informative sampling designs. After a run of releases from 2019 to 2022, the feed goes quiet until this year's 1.1.0, which breaks the control interface, exposes the block split and merge operations as testable instance methods, and adds a node-swap refinement pass that runs after variational convergence.
2026.8.0 is out, closing a seven-candidate train that ran through early August. The GA tag itself is the last cherry-pick batch — SCIM group membership removals, a proxy redirect that preserves query strings, session deletion on user deactivation — but the release it finalizes is where the substance lives: an Actors primitive in core, an enterprise Agent requiring a domain join and its own API scope, OAuth2 token exchange delegation, and a CAS source integration.
Two threads converge in this major. The identity surface keeps broadening at the protocol edge — CAS, WS-Fed, token exchange delegation, on-behalf-of — while the enterprise tier grows an endpoint story that reaches machines and devices rather than browser sessions. The RC train's shape reinforces it: six candidates fired in one day on CI and docs, then one heavy candidate carrying the features, then a fix-only close. That is release engineering hardened around a major, not a routine point release.
With Agents and Actors now GA rather than cherry-picks, the next branch should build out what they enable — device-conditioned policies or agent-brokered credentials — while 2026.8.x settles into backport patches.
The package fits stochastic block models to networks with missing data, covering both missing-at-random and informative sampling designs. After a run of releases from 2019 to 2022, the feed goes quiet until this year's 1.1.0, which breaks the control interface, exposes the block split and merge operations as testable instance methods, and adds a node-swap refinement pass that runs after variational convergence.
The new release is maintenance-driven in the best sense: it targets the parts of the codebase that were hard to test or easy to misuse. Replacing free-form control lists with a function of named, defaulted arguments turns silent typos into errors, and pulling the exploration logic out of the collection class makes the search algorithm independently testable without changing it. The polish step addresses a known weakness, reaching individually misclassified nodes that split and merge moves cannot fix.
Given the gap before this release, the near-term question is whether the cadence resumes at all; the refactoring it contains would support further algorithmic work if it does.
Other Infra & APIs products tracked by Sparkpulse, ranked by recent ship velocity. Each card links to a full editorial trajectory and lets you pivot into a head-to-head comparison with either authentik or missSBM.
Observability lands on OpenTelemetry semconv in the LTS train
Canvas agents gain memory, and onboarding moves into the editor
Security and governance controls catch up to the Copilot build-out
Rancher's public feed is a build-tag stream: three branches bumped the same Go image on one afternoon
Buildkite keeps converting hand-rolled agent workarounds into first-class CI primitives.
Cursor's agents stop waiting to be asked - they subscribe, and they hold a goal until it's done.
See all authentik alternatives → · See all missSBM alternatives →
Latest ship moves from both products, interleaved chronologically. ⚡ = editorial spark.
They serve adjacent needs but don't currently overlap on shipped themes. authentik is currently shipping more aggressively (velocity 6.3 vs 2.5), with 1 editorial sparks in the last 30 days against 0. See the at-a-glance table above for a side-by-side breakdown of velocity, recent sparks, and editorial themes.
Sparkpulse doesn't pick a winner — we score release velocity, not feature parity. authentik is currently shipping more aggressively (velocity 6.3 vs 2.5), with 1 editorial sparks in the last 30 days against 0. For your specific use case, the alternatives sections above list other Infra & APIs products to evaluate alongside.
Top authentik alternatives in Infra & APIs are ranked by recent ship velocity. Browse the "authentik alternatives" section above for the current picks, or visit /alternatives/authentik for the full list with editorial commentary on each.
Top missSBM alternatives in Infra & APIs are ranked by recent ship velocity. Browse the "missSBM alternatives" section above for the current picks, or visit /alternatives/misssbm for the full list with editorial commentary on each.