Wowza
Wowza keeps arguing its case for detection inside the streaming pipeline
A side-by-side editorial comparison of 3CX and FreeSWITCH — release velocity, themes, recent moves, and the top alternatives to consider.
3CX is rebuilding around an AI Server, and it just handed the channel the AI edition to sell.
Recent releases split between the AI Server — sizing guidance, deployment updates, MCP and admin control in the V20 Update 10 alpha — and steady client work, most recently the V5.8 softphone beta with one-click SSO and outbound number choice. Packaging moved too: partner and distributor NFR subscriptions were replaced with the AI Edition.
With the legacy purge done, FreeSWITCH spent 1.11.2 bounds-checking everything that parses bytes
FreeSWITCH shipped 1.11.2 in August 2026, three months after the 1.11.0 branch cut that removed roughly thirty modules and migrated the regex engine to PCRE2. Where 1.11.0 was about deleting surface area, 1.11.2 is about auditing what remains: nearly every bug line is a bounds or length check in a parser — STUN attributes, base64 decode, XML entity decoding, RTMP H.264 NAL units, XML-RPC WebSocket frames, MSRP body lengths. Alongside the fixes it adds an interface allowlist and turns on DTLS client-certificate verification against the SDP fingerprint plus opt-in STUN MESSAGE-INTEGRITY checks on inbound ICE. The build system continues moving off self-hosted infrastructure onto GitHub-published artifacts and upstream packages.
Recent releases split between the AI Server — sizing guidance, deployment updates, MCP and admin control in the V20 Update 10 alpha — and steady client work, most recently the V5.8 softphone beta with one-click SSO and outbound number choice. Packaging moved too: partner and distributor NFR subscriptions were replaced with the AI Edition.
The AI Server is being treated as infrastructure customers deploy themselves rather than a hosted add-on, with hardware guidance, dedicated deployment tooling and an MCP endpoint for connecting assistants. Putting the AI Edition into partner hands and easing self-managed subscription changes both point at pushing that stack out through the channel rather than upselling it centrally.
Expect V20 Update 10 to move from alpha toward release with the AI Server management changes already previewed, and the V5.8 softphone beta to reach general availability.
FreeSWITCH shipped 1.11.2 in August 2026, three months after the 1.11.0 branch cut that removed roughly thirty modules and migrated the regex engine to PCRE2. Where 1.11.0 was about deleting surface area, 1.11.2 is about auditing what remains: nearly every bug line is a bounds or length check in a parser — STUN attributes, base64 decode, XML entity decoding, RTMP H.264 NAL units, XML-RPC WebSocket frames, MSRP body lengths. Alongside the fixes it adds an interface allowlist and turns on DTLS client-certificate verification against the SDP fingerprint plus opt-in STUN MESSAGE-INTEGRITY checks on inbound ICE. The build system continues moving off self-hosted infrastructure onto GitHub-published artifacts and upstream packages.
The 1.11.x line reads as a deliberate two-step: cut the maintenance surface, then harden what is left. Having removed thirty modules, the project can now afford to audit the media and signaling paths line by line, and the pattern of fixes — untrusted length fields reaching memory writes — suggests a systematic pass rather than incremental reports. The parallel thread is dependency modernization: mod_v8 moved from a custom v8-6.1 fork to upstream libnode, SpanDSP and libvpx were bumped, and Windows builds now pull pre-compiled FFmpeg and libcodec2 rather than building them. Both threads point the same direction — less bespoke code to own.
Expect the hardening pass to continue into the remaining parsers, with the opt-in STUN MESSAGE-INTEGRITY check becoming default once operators have shipped through a release. The build-system migration off files.freeswitch.org appears partway done, so the next release likely finishes moving the remaining assets to GitHub-hosted artifacts.
Other Meetings 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 3CX or FreeSWITCH.
Wowza keeps arguing its case for detection inside the streaming pipeline
Vimeo ships another batched product post: six changes to embeds and the player
Switcher Studio's feed is a livestreaming how-to blog, not a changelog
Every manual control Mux ships now arrives with a Robots workflow that does it for you
WebinarJam's feed is a funnel-marketing curriculum with no product releases in it.
Muvi cleared SOC 2 and immediately turned it into the centerpiece of its enterprise pitch.
See all 3CX alternatives → · See all FreeSWITCH alternatives →
Latest ship moves from both products, interleaved chronologically. ⚡ = editorial spark.
Both compete on the same themes — voip — within Meetings. 3CX is currently shipping more aggressively (velocity 6.3 vs 2.5), with 0 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. 3CX is currently shipping more aggressively (velocity 6.3 vs 2.5), with 0 editorial sparks in the last 30 days against 0. For your specific use case, the alternatives sections above list other Meetings products to evaluate alongside.
Top 3CX alternatives in Meetings are ranked by recent ship velocity. Browse the "3CX alternatives" section above for the current picks, or visit /alternatives/3cx for the full list with editorial commentary on each.
Top FreeSWITCH alternatives in Meetings are ranked by recent ship velocity. Browse the "FreeSWITCH alternatives" section above for the current picks, or visit /alternatives/freeswitch for the full list with editorial commentary on each.