Fish Audio Live 3 vs Grok Voice 3.0: un precio es un paquete y el otro es un componente

Put the two rate cards side by side and the newer model looks less than half the price. Fish Audio Live 3 sells for $0.035 per minute on Fish Audio's Agent Platform — a promotional rate, half of a $0.07 list, good through 31 December 2026. G​rok Voice 3.0 sells for $0.08 per minute, or $4.80 an hour, plus $0.004 per text input token. Four cents against eight is a clean story until you notice that the two numbers are billing different objects: G​rok Voice 3.0 is a speech-to-speech component you rent, and Live 3's per-minute figure bundles speech recognition, the reasoning model and the voice into one meter. One is a part, the other is an assembly. Compare them per minute and you are comparing a gearbox to a car.

That distinction is the whole comparison, and it is worth being precise about why. G​rok Voice 3.0 is a versioned model string behind a realtime WebSocket with session-scoped tools and returned transcripts, priced by the minute of audio it processes and scored, independently, at 86.9 on the Artificial Analysis Speech to Speech Index. Fish Audio Live 3 is a full-duplex model launched on 10 November 2026 that keeps listening while it speaks, is English-only at launch by the company's own statement, has no published benchmark of any kind, no model identifier in Fish Audio's API documentation, and no weight release — the vLLM project's public reply to the launch was to hope it gets open-sourced one day.

El marcador que existe solo en un lado

Artificial Analysis measures G​rok Voice 3.0 on the same harness it uses for every duplex model, and the numbers are unusually complete. Every figure below is the board's own, taken from its run dated 13 October 2026.

• Razonamiento del habla, Big Bench Audio — 98,5 %.

• Dinámica conversacional, Full Duplex Bench — 97,0%, con backchannelling 92,9%, gestión de interrupciones 96,5%, toma de turnos 100% y gestión de pausas 98,5%.

• Speech Agent Arena preference — Elo 1059, against 1074 for G​emini 3.8 Live and 990 for Eleven Realtime v1 Max.

• Éxito en la tarea — 94,8 % en los escenarios de aerolíneas, comercio minorista y telecomunicaciones de τ-Voice.

• Time to first audio — 0.81 seconds, slower than G​rok Voice 2.0's 0.70 and roughly 40% quicker than GPT-Live-1 in its Astra medium configuration at 1.34 seconds.

• Costo por hora de audio de entrada: $4.80, una conversión directa de la tarifa por minuto del propio proveedor.

Live 3's column is not sparse, it is empty, and the emptiness has a specific shape. There is no row on that board for any Fish Audio duplex model as of 11 November 2026 — the board lists Fish Audio as a model family and scores none of its live work. Fish Audio's one published number is 186 milliseconds to decide to respond, which the company describes in the context of short reactions such as "mm-hmm" rather than full answers. It is a vendor floor on an acknowledgement. G​rok's 0.81 seconds is an audited time-to-first-audio on a reply. Nothing about those two figures is commensurable, and the fact that both are expressed in seconds is the trap.

A qué se compromete realmente la afirmación de dúplex de cada parte

Ambos modelos son full duplex, así que «escucha mientras habla» no es un diferenciador entre ellos; es el requisito de entrada. La diferencia está en de dónde procede la política de temporización.

El argumento de Fish Audio es que el acuse, la espera, la respuesta y la detención se aprendieron de conversaciones humanas grabadas en lugar de escribirse como una lista de umbrales de silencio, y lo ilustra con un número de teléfono leído por fragmentos: un agente que oye tres eventos de fin de turno donde una persona oye una sola vacilación es el fallo que se está corrigiendo. El material de apoyo son tres llamadas de ejemplo: finalizar un pedido, interrumpir y emitir señales de seguimiento durante una lectura de confirmación. Lo que Fish Audio no ha publicado es ninguna medición de si la política aprendida supera a una basada en umbrales.

G​rok Voice 3.0's timing policy is not explained in behavioural terms at all. What xAI published is a rate card, a model string, an alias move — grok-voice-latest now resolves to grok-voice-3.0, and grok-voice-2.0 is flagged deprecated — and, through Artificial Analysis, a set of turn-taking scores that a buyer can read as a risk statement. A 96.5% interruption score is a claim you can test against your own call recordings. A 186-millisecond acknowledgement floor is a claim you have to take on trust until you run it.

The alias move matters more than it sounds in a phone system. G​rok Voice 2.0 became the alias target on 5 August 2026 and was superseded by 3.0 ten weeks later, which means a team that deployed against the alias has already been through two silent model swaps in a quarter. Whether you pin grok-voice-3.0 explicitly is a deployment decision the vendor has effectively forced.

Dónde acaba realmente el dinero

La división entre paquete y componente genera cuatro perfiles de coste, y solo uno de ellos favorece a Live 3 de una manera que resiste el contacto con una llamada real.

• Live 3 en Agent Platform — $0.035 por minuto todo incluido durante la promoción, cubriendo reconocimiento, razonamiento y voz; añada $0.01 por minuto para llamadas telefónicas, $1.20 al mes por número de teléfono, y la tarifa por token del LLM de nivel premium si actualiza desde el modelo incluido. Una llamada de diez minutos cuesta 35 centavos más la telefonía.

• Live 3 after 31 December 2026 — the list rate is $0.07 per minute, at which point the all-in figure is within a cent of G​rok's $0.08 and the bundle argument has to carry the comparison on its own. Fish Audio has not said what happens to the discount.

• G​rok Voice 3.0 — $0.08 per minute of audio plus $0.004 per text input token, with the reasoning model, the telephony and the orchestration all bought separately. The per-minute rate is a floor, not a total.

• O bien, con un modelo de razonamiento propio —la única configuración en la que los dos convergen, porque ambos proveedores te permiten aportar el cerebro y facturan la capa de voz.

Ese último punto es la conclusión práctica que se esconde dentro de la tabla de precios. Si tu producto se diferencia en la capa de razonamiento, el paquete es un descuento sobre algo que ibas a reemplazar. Si no, el paquete es genuinamente más barato y sencillo, y la promoción del 50% es la mejor oferta en voz este mes.

Compras, y la pregunta que ningún tarifario responde

El precio y las puntuaciones son la mitad fácil. La mitad difícil es lo que ocurre cuando el modelo cambia bajo tus pies, y aquí los dos proveedores han hecho apuestas opuestas.

G​rok Voice 3.0 has a stable interface and an unstable model: same endpoint, same event names, same bearer-token handshake, same voice set, with the model swapped behind an alias. Fish Audio Live 3 has an unstable interface and, for now, a stable model: the platform is in public beta, the duplex model is new, its API documentation does not yet name it, and the launch post is the only document that describes its behaviour. One arrangement asks you to test continuously. The other asks you to read release notes.

Ninguna de las dos es irrazonable. Pero un equipo de ingeniería que elige entre ellas debería fijarse en qué riesgo está mejor equipado para absorber, porque esa respuesta suele ser más predictiva de un despliegue exitoso que cuatro centavos por minuto.

La capa portátil y el límite honesto

Both stacks are two-part systems, and in both the speech half is unportable — G​rok Voice 3.0 runs only on xAI's realtime endpoint, Live 3 only on Fish Audio's platform and on enterprise hardware. The reasoning half is where a router belongs, and on both it is a separate purchase: on the G​rok side it is a model you call from inside the session's tools, on the Fish side it is an OpenAI-compatible endpoint the agent points at. OrcaRouter carries more than 200 models behind one key at provider list price with zero markup, so the text model in either stack can be swapped or pinned without renegotiating the speech contract, and automatic failover keeps a session alive when the reasoning backend errors mid-call. Neither speech model is routable here — there is no third-party endpoint for either — and this piece would be misleading if it implied otherwise.

Para un equipo que aún está decidiendo, el experimento más barato es el que permite el esquema de precios: ejecutar las mismas veinte llamadas en ambos, puntuar tú mismo las interrupciones y ver si el 94,8 % de éxito en tareas en los escenarios minoristas de otra persona predice algo sobre los tuyos. Los 186 milisegundos de Live 3 se sentirán como la razón de ser del modelo o como una cifra sobre un ruido que el agente produce mientras piensa. Esa no es una cuestión que una tarifa pueda resolver, y es la única cuestión que importa cuando llega la factura.