ElevenLabs dropped from LiveKit Inference as Rime TTS voices go free
LiveKit removed ElevenLabs entirely from its Inference catalog (all STT and TTS SKUs) and cut Rime's four TTS voices to $0.00/min on every Cloud plan; no Cloud plan price changed.
TTS catalog included six ElevenLabs voices (Eleven Flash v2, Flash v2.5, Multilingual v2, Turbo v2, Turbo v2.5, Eleven v3) and STT catalog included ElevenLabs Scribe v2 Realtime ($0.0105/min); Rime's four voices (Coda, Mist, Mist v2, Mist v3) were priced $50.00/$30.00/$30.00/$30.00 per million characters on Build/Ship and $50.00/$20.00/$20.00/$20.00 on Scale; Cloud's metered table had no 'Non-production deployments' row.
ElevenLabs is delisted entirely from both the STT and TTS catalogs (zero ElevenLabs SKUs remain). Rime's four voices are now Free ($0.00) on both Build/Ship and Scale. Two new LLM SKUs join the catalog — Google Gemini 3.8 Flash ($0.0029/min) and xAI/SpaceXAI Grok 4.6 ($0.0070/min) — plus new TTS entries Cartesia Sonic 3.6 and Inworld Realtime TTS 2.0 Flash. LiveKit Cloud's metered-units table also gains a new 'Non-production deployments' row (Build 0, Ship 2, Scale 5, Enterprise Custom). Build ($0), Ship ($50/mo), Scale ($500/mo) and Enterprise (custom) plan prices are unchanged.
LiveKit’s per-minute Inference catalog — the single-API-key service for calling LLM, STT and TTS models against each Cloud plan’s credit allotment — lost an entire vendor this week. ElevenLabs, which supplied one speech-to-text model (Scribe v2 Realtime) and six text-to-speech voices (Eleven Flash v2, Flash v2.5, Multilingual v2, Turbo v2, Turbo v2.5 and Eleven v3), no longer appears anywhere in the STT or TTS pricing tables on either the main pricing page or the dedicated Inference pricing page.
At the same time, Rime’s entire four-voice lineup (Coda, Mist, Mist v2, Mist v3) moved from paid ($50.00/M characters for Coda, $30.00/M for the three Mist voices on Build/Ship; $50.00/$20.00 on Scale) to Free ($0.00) on every plan tier — joining Fish Audio’s S2.1 Pro Free and Deepgram’s Flux TTS as free SKUs already on the catalog, and bringing the total count of $0/min TTS models to six.
The catalog also picked up two new LLM entries (Google Gemini 3.8 Flash at $0.0029/min and a sixth xAI/SpaceXAI Grok model, Grok 4.6, at $0.0070/min) and two new TTS SKUs (Cartesia Sonic 3.6 and Inworld Realtime TTS 2.0 Flash), while a duplicate AssemblyAI SKU (Universal-3 Pro Streaming, superseded by Universal-3.5 Pro Streaming) was cleaned up. Separately, LiveKit Cloud’s metered-units table gained a new “Non-production deployments” allotment (Build 0, Ship 2, Scale 5, Enterprise Custom) alongside existing agent-deployment limits. No LiveKit Cloud plan price, and no other allotment or overage rate, moved: Build remains free, Ship starts at $50/mo, Scale at $500/mo, and Enterprise stays custom-quoted.
Plan structure holds (Build free / Ship $50 / Scale $500 / Enterprise custom). LiveKit Inference delists ElevenLabs entirely (1 STT + 6 TTS SKUs) and cuts Rime's four TTS voices to $0.00/min on every tier; adds Google Gemini 3.8 Flash and SpaceXAI Grok 4.6 (LLM) plus Cartesia Sonic 3.6 and Inworld Realtime TTS 2.0 Flash (TTS); removes a duplicate AssemblyAI STT SKU; and Cloud's metered-units table gains a new 'Non-production deployments' allotment row (Build 0 / Ship 2 / Scale 5 / Enterprise Custom), a per-tier cap with no overage rate.