Krisp
Krisp adds deepfake detection and expands Voice Translation as its Call Center AI platform deepens enterprise compliance tools.
A side-by-side editorial comparison of Dendrite and Rocket.Chat — release velocity, themes, recent moves, and the top alternatives to consider.
A forked, relicensed Dendrite spent its first year catching up on Matrix protocol and security debt
The window opens with the first release after forking from matrix-org/dendrite under AGPLv3, which fixed long-standing state-reset bugs and added I2P and Onion listening. What follows is a security release closing a server-side request forgery vulnerability, then a release adding room version 12 support and two MSC implementations, and finally two quick patches for unresponsiveness, v12 room joins and malformed PDUs blocking whole transactions.
Rocket.Chat finishes the 8.8 calling work while the licence wall for SSO gets a date.
The 8.8 candidate line is winding down. rc.2 is a bare Meteor version bump with no user-visible change, following rc.1 which promoted Voice Call Screen Sharing to stable and turned on the in-room call interface for every call, screen sharing available or not, alongside repairs to a message list that could quietly load a whole room's history in the background. The release that still defines the line is rc.0: LDAP and SAML deprecated on non-Premium workspaces with 9.0.0 named as the cutoff, forced end-to-end encryption for new private rooms, classification banners for ABAC-managed rooms, and the apps-engine default runtime moved from Deno to Node.
The window opens with the first release after forking from matrix-org/dendrite under AGPLv3, which fixed long-standing state-reset bugs and added I2P and Onion listening. What follows is a security release closing a server-side request forgery vulnerability, then a release adding room version 12 support and two MSC implementations, and finally two quick patches for unresponsiveness, v12 room joins and malformed PDUs blocking whole transactions.
This is a fork proving it can carry a homeserver. The order is telling: license and state correctness first, then a CVE, then protocol currency with room version 12, then the operational bugs that surface once people actually run it. Several fixes carry outside contributor credits, which is the fork's clearest sign of life.
Expect continued protocol catch-up on newer MSCs and room versions plus operational fixes, since federation compatibility is the constraint that sets this project's agenda.
The 8.8 candidate line is winding down. rc.2 is a bare Meteor version bump with no user-visible change, following rc.1 which promoted Voice Call Screen Sharing to stable and turned on the in-room call interface for every call, screen sharing available or not, alongside repairs to a message list that could quietly load a whole room's history in the background. The release that still defines the line is rc.0: LDAP and SAML deprecated on non-Premium workspaces with 9.0.0 named as the cutoff, forced end-to-end encryption for new private rooms, classification banners for ABAC-managed rooms, and the apps-engine default runtime moved from Deno to Node.
Two threads run at once and point in different directions. The calling stack keeps maturing on its own — voice, then screen sharing, now a consistent in-room surface for both — which is Rocket.Chat absorbing functions its users would otherwise reach for a separate tool to get. Against that, the 9.0 licence requirement for LDAP and SAML sorts the install base: regulated and government deployments that will pay, and unlicensed self-hosted workspaces that were there for the absence of a per-seat bill. The rc train itself has settled into a familiar shape, one substantive rc.0 followed by bumps and fixes.
Expect 8.8 to go final once the message-list fixes settle, with the LDAP and SAML licence requirement, not any calling feature, as the story of 9.0. The open question is whether the Deno-to-Node runtime switch surfaces app compatibility reports before then.
Other Comms 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 Dendrite or Rocket.Chat.
Krisp adds deepfake detection and expands Voice Translation as its Call Center AI platform deepens enterprise compliance tools.
Netcore's recent entries are blog content, not product releases — the signal is a BFSI and ecommerce vertical sales push.
Canary Mail ships AI chat history while sustaining a cross-platform bug fix cadence.
Superhuman is becoming the email layer for AI agents, not just the fastest email client.
Salesmsg now runs the entire sales-to-NPS loop in one thread — outbound, AI reply, booking, survey, and routing.
Zoho Cliq 7.0 embeds AI task extraction and ships a revamped external collaboration layer
See all Dendrite alternatives → · See all Rocket.Chat alternatives →
Latest ship moves from both products, interleaved chronologically. ⚡ = editorial spark.
They serve adjacent needs but don't currently overlap on shipped themes. Rocket.Chat is currently shipping more aggressively (velocity 6.3 vs 0.0), 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. Rocket.Chat is currently shipping more aggressively (velocity 6.3 vs 0.0), with 1 editorial sparks in the last 30 days against 0. For your specific use case, the alternatives sections above list other Comms products to evaluate alongside.
Top Dendrite alternatives in Comms are ranked by recent ship velocity. Browse the "Dendrite alternatives" section above for the current picks, or visit /alternatives/dendrite for the full list with editorial commentary on each.
Top Rocket.Chat alternatives in Comms are ranked by recent ship velocity. Browse the "Rocket.Chat alternatives" section above for the current picks, or visit /alternatives/rocketchat for the full list with editorial commentary on each.