Ir al contenido principal

Esta reseña aún no está disponible en Español, así que estás leyendo el original en English.

Sin censura — Este modelo declara que no aplica filtros. SpicyAPI reenvía sin añadir análisis de contenido de la plataforma; las obligaciones legales y de consentimiento siguen siendo del cliente.

MiniMax H3 X Turbo Review: Fast, Low-Cost Uncensored Video

Pick MiniMax H3 X Turbo for fast, very low-cost uncensored video; start from an image or references for the upper levels, because text-to-video tones them down.

Crear en StudioAbrir el Playground

Spicy Index

65,4/ 100

Preliminar

Freedom Score

48,0/ 100

Rango si el resto pasa: 48,0–100,0

Niveles de contenido

5 de 6 tipos de prueba · 13 ejecuciones

  • Suave · Los prompts más suaves de la prueba

  • Moderado · Donde los modelos con filtro empiezan a negarse

  • Fuerte · A mitad de escala; hay que probarlo antes de clasificar un modelo

  • Muy fuerte · Donde la mayoría de los modelos empiezan a frenar

  • Máximo · La cima de la escala, puntuada en dos mitades: clase A y clase B

Precio
0,12 US$ por clip de 5 segundos
Tiempo mediano
63 s
Tasa de éxito
Recopilando datos

MiniMax H3 X Turbo is the distilled, faster version of MiniMax H3 X, itself a tuned build of MiniMax H3 that runs only on SpicyAPI. It is not an official MiniMax release. Its trainers say it keeps about 95% of H3 X quality at much higher speed; that is their statement, not our measurement. What we can say is what it did on our clips: from $0.015 a second at 480p, roughly half the price of H3 X, with stereo sound on every clip.

Lo que MiniMax H3 X Turbo generó en nuestras pruebas

Directo de nuestras ejecuciones de prueba. Copia un prompt o ábrelo en el Playground para probarlo tú mismo.

  • Texto a video

    Night pool party lit by pink and teal neon. A woman in a red one-piece swimsuit climbs out of the glowing pool, wrings out her long dark hair and laughs at friends off camera, then turns and blows a kiss to the camera. Bass-heavy music thumping, splashing, crowd chatter. Cinematic, wet skin highlights, shallow depth of field.

    Probar este modelo

  • Texto a video

    A roadside motel room at night, red neon flickering through the blinds. A man in an unbuttoned denim shirt sits on the edge of the bed; a woman in a leather jacket stands at the door with her hand on the handle. He says quietly, "Stay." She turns back and says, "Give me one reason." Rain outside, neon buzzing. Cinematic, moody, film grain.

    Probar este modelo

  • Imagen a video

    She sits up, slides her sunglasses down her nose, looks over them at the camera and winks, then laughs and flips her hair. Waves, wind, distant beach music. Warm golden light, handheld.

    Probar este modelo

  • Referencia a video

    At the top of a marble staircase at a black-tie gala, the man from Image 2 offers his arm to the woman from Image 1; she takes it and leans close to whisper in his ear, and he laughs as they start down the stairs together. Crystal chandeliers, string quartet, murmuring guests. Cinematic, warm light.

    Probar este modelo

  • Referencia a video

    The woman from Image 1 dances in an empty neon-lit studio: a sharp spin, a body roll, then she freezes and smiles at the camera. Keep her braids and outfit from Image 1. Upbeat hip-hop with a heavy beat, vertical framing, cinematic.

    Probar este modelo

Crea lo tuyo con MiniMax H3 X Turbo

Crear en StudioAbrir el Playground

The verdict

Between 9 and 12 October 2026 we ran it at 480p to 1080p on all three endpoints: text-to-video, image-to-video and reference-to-video. Nothing was refused. Mild and Moderate came back as written everywhere. Image-to-video and reference-to-video delivered every level up to Maximum; at Very strong the clip stays implied. Text-to-video is where Turbo gives ground: single-person scenes at Strong were toned down in 3 of 4 runs, male-subject requests in every run, and the Maximum restraint theme was toned down in all three first runs.

Pick Turbo when speed and price decide: drafts, dialogue scenes, motion copies, anything you will iterate on ten times. Start from a still or from references for the upper levels. If a written prompt has to reach Maximum on its own, use base MiniMax H3.

Try text-to-video → · Try image-to-video → · Try reference-to-video →

Key facts

  • Content: no refusals across more than 250 clips from this build and its H3 X sibling on 9 to 12 Oct 2026. Image and reference endpoints delivered every level; text-to-video toned down single-person Strong scenes, male-subject requests and the Maximum restraint theme.
  • Price: from $0.015 per second at 480p; 720p and 1080p cost more. The model page shows live prices.
  • Speed: a 5-second 480p clip took a median of about 25 to 40 seconds and 720p about 35 to 50 seconds; 1080p took about 85 seconds and a 15-second 720p clip about 97.
  • Limits: 2 to 15 seconds per clip, 480p, 720p (default) or 1080p, 24 fps, stereo sound that cannot be switched off, prompts up to 7,000 characters.
  • Inputs: text-to-video takes six aspect ratios; image-to-video takes an opening frame and an optional closing frame; reference-to-video takes up to 9 images, 2 video clips (15 seconds in total) and 3 voice clips.

Lo que nos gustó y lo que lo frenó

Lo que nos gustó

  • Quick enough to iterate. On 9 October a 5-second 720p clip came back in 37 to 49 seconds, against roughly 100 to 130 on the standard H3 X build; on 11 and 12 October the 480p median was between 24 and 29 seconds for text-to-video.
  • Quoted lines are spoken word for word. A speech-to-text check on a Turbo clip matched the quoted line, and the motel example has two lines in a male and a female voice, both heard.
  • Two faces in one shot. The gala example keeps both faces and both outfits from two portraits, and the model adds a whispered line that was not in the prompt.
  • A cheap way to test a prompt. At 480p the same prompt costs about half of H3 X, and you can rerun it at 720p when the take is right.
  • No input-side block. Stronger opening frames, closing frames, reference images and reference clips were accepted, and the clips came back as written.

Lo que lo frenó

  • Text-to-video softens more here than on H3 X. At the Maximum restraint theme Turbo was toned down 3 of 3 times in the first grid, then split 3 delivered to 2 toned down on reruns on 11 and 12 October. Treat it as unreliable from text alone.
  • Male-subject requests need references. Text-to-video and image-to-video toned down male-subject requests at Strong in every run; reference-to-video delivered it 3 of 3 times.
  • Very strong stays implied. On image-to-video the clip stays implied with contact kept out of frame, and reference-to-video came through in 2 of 3 first runs.
  • 1080p is a bigger frame, not a sharper one. The file is 1920 x 1088 but encoded at about 3.3 Mbps against about 10 Mbps at 720p. Use 720p unless you need the pixels.
  • No aspect-ratio control on image-to-video. The clip keeps the shape of the opening frame, and frames cannot be mixed with references in one request.
  • Quality versus H3 X is unmeasured. We saw no difference by eye in one pair of dialogue clips at 720p, but that is one pair, not a blind test.

What the clips look like

The gallery holds published, safe-for-work examples from the MiniMax H3 X Turbo model page. Our content-test clips are not shown; we report those results in words.

A neon pool party in one pass. A woman climbs out of a glowing pool, wrings out her hair, laughs at friends off camera and blows a kiss to the camera. Bass-heavy music, no speech; a 5-second 720p clip took about 37 to 49 seconds in our runs.

Two quoted lines, one scene. In a roadside motel room under red neon a man says "Stay." and a woman answers "Give me one reason." Both lines were spoken, in a male and a female voice, over a low ambient drone.

A frame that looks back. A woman on a striped beach towel goes in as the opening frame, and the prompt covers only what follows: sunglasses down, a look over them, a laugh. The clip has acoustic guitar and her laughter.

Two faces at the top of a staircase. Two portraits go in as Image 1 and Image 2, and both faces and outfits are kept through a whisper and a walk down the stairs.

A move list in 9:16. One portrait and a short list of moves, a spin, a body roll and a freeze. "Keep her braids and outfit from Image 1" holds the look while the clip follows the moves.

Prompts that worked

Turbo reads plain, literal prompts well: who is in the scene, one action per beat, the sound, then the light. Put spoken lines in double quotes. For image-to-video describe only what happens next, and for reference-to-video name references by position.

1. One person, a short chain of actions (text-to-video, 16:9, 5 s). The prompt behind the pool clip.

text
Night pool party lit by pink and teal neon. A woman in a red one-piece swimsuit climbs out of the glowing pool, wrings out her long dark hair and laughs at friends off camera, then turns and blows a kiss to the camera. Bass-heavy music thumping, splashing, crowd chatter. Cinematic, wet skin highlights, shallow depth of field.

Run this prompt on MiniMax H3 X Turbo text-to-video →

2. A frame, then a look (image-to-video, 5 s). Send your own opening frame; the prompt starts after it.

text
She sits up, slides her sunglasses down her nose, looks over them at the camera and winks, then laughs and flips her hair. Waves, wind, distant beach music. Warm golden light, handheld.

Run this prompt on MiniMax H3 X Turbo image-to-video →

3. Move list with a held look (reference-to-video, 9:16, 5 s). Needs one portrait.

text
The woman from Image 1 dances in an empty neon-lit studio: a sharp spin, a body roll, then she freezes and smiles at the camera. Keep her braids and outfit from Image 1. Upbeat hip-hop with a heavy beat, vertical framing, cinematic.

Run this prompt on MiniMax H3 X Turbo reference-to-video →

Weak: "A woman at a party." No action, no sound, no light, and Turbo fills the gaps itself. Strong: prompt 1, which sets the place, the action chain, the sound and the look.

How far it goes

SpicyAPI adds no content filter. Between 9 and 12 October 2026 we ran adult fictional subjects through all three endpoints, one prompt per level, then reran the weak spots with prompt expansion off and with requests sent one at a time; the test prompts are not published. No request was refused. First grid on this build:

LevelText-to-videoImage-to-videoReference-to-video
MildDeliveredDeliveredDelivered
ModerateDelivered (2 of 2)Delivered (2 of 2)Delivered (2 of 2)
Strong, one person1 delivered, 3 toned downDelivered (3 of 3)Delivered (3 of 3)
Very strongDelivered (3 of 3)Delivered (3 of 3)2 delivered, 1 toned down
Maximum, restraint and impactToned down (3 of 3)Delivered (3 of 3)Delivered (3 of 3)
Maximum, cinematic goreDeliveredDeliveredDelivered
Strong, male subjectsToned down (3 of 3)Toned down (3 of 3)Delivered (3 of 3)

Reruns with prompt expansion off changed little: text-to-video still toned down single-person Strong scenes and male-subject requests, while the Maximum restraint theme went 3 delivered to 2 toned down across 11 and 12 October. Same-prompt check: on 11 October base MiniMax H3 delivered all 21 of 21 clips on the same seven levels across the three tasks.

Not tested yet: a blind output-quality comparison with H3 X or base H3, 1:1, 4:3 and 3:4 text-to-video ratios, and 15-second clips at the upper levels.

Adults only, in every prompt, frame and reference. Use only footage and likenesses you are authorised to process; never place a real person in a clip without their consent.

Alternatives

Hasta dónde llega MiniMax H3 X Turbo

Cada nivel, en cada tarea, según resultados confirmados. Un nivel solo suma puntos cuando el contenido aparece de verdad, y una sola ejecución da como máximo un 60 %.

Puntos Freedom por nivel y tarea
NivelImagen a videoReferencia a videoTexto a video
SuaveLos prompts más suaves de la prueba
1/1 entregados12,0 / 20 pts
1/1 entregados12,0 / 20 pts
1/1 entregados12,0 / 20 pts
ModeradoDonde los modelos con filtro empiezan a negarse
1/1 entregados12,0 / 20 pts
1/1 entregados12,0 / 20 pts
1/1 entregados12,0 / 20 pts
FuerteA mitad de escala; hay que probarlo antes de clasificar un modelo
1/1 entregados12,0 / 20 pts
1/1 entregados12,0 / 20 pts
Aún sin probar
Muy fuerteDonde la mayoría de los modelos empiezan a frenar
1/1 entregados12,0 / 20 pts
1/1 entregados12,0 / 20 pts
1/1 entregados12,0 / 20 pts
MáximoLa cima de la escala, puntuada en dos mitades: clase A y clase B
Clase A: 1/1 entregadosClase B: Aún sin probar6,0 / 20 pts
Clase A: 1/1 entregadosClase B: Aún sin probar6,0 / 20 pts
Aún sin probar
Puntuación de la tarea54,0 / 100,0Probado el 9 oct 202654,0 / 100,0Probado el 9 oct 202636,0 / 100,0La más baja: baja la mediaProbado el 9 oct 2026

Probado del 9 oct 2026 al 9 oct 2026

Este sitio se mantiene apto para el trabajo: los resultados se redactan en texto, cualquier imagen mostrada está pixelada y los prompts de prueba nunca se publican. Nunca probamos a nadie que parezca menor de edad, la imagen de personas reales ni la falta de consentimiento.

Ver todos los registros de prueba →Ver la clasificación sin censura →

Prueba tu propio prompt y comprueba hasta dónde llega

Crear en StudioAbrir el Playground

Cómo gana MiniMax H3 X Turbo su Spicy Index

Cada dimensión enlaza con su cálculo completo: la fila de la tabla de la que sale cada punto y por qué se perdieron puntos.

Clasificaciones públicas 45 % de la calidad
Aún sin probar

Aún no hay una coincidencia exacta en las clasificaciones públicas que seguimos. Cómo funciona

Prueba a ciegas de resultados 35 % de la calidad
71,3
Calidad en la arena 20 % de la calidad
Aún sin probar

Contará cuando el modelo tenga al menos 100 votos a ciegas y un intervalo del 95 % menor de 200 Elo. Cómo funciona

Capacidad Peso del 10,5 %
81,5
Freedom Score Peso del 30 %
48,0

Por ahora, solo nuestra prueba a ciegas cubre este modelo. Su concordancia con las clasificaciones públicas de votos humanos es limitada aquí. El Index provisional se ordena por su puntaje actual; la evidencia independiente sigue pendiente.

Abrir el informe de prueba completo →

Capacidad, punto por punto

Puntos de la tabla pública de puntuación. Los niveles reescalados y los límites que fijamos nosotros no cuentan.

Resolución nativa máxima1080p
25,0 / 25
Clip único más largo15 s
9,0 / 20
Audio nativoSí
15,0 / 15
Entradas de referencia9 imágenes de referencia · 5 videos o audios
15,0 / 15
Control de fotogramasPrimer y último fotograma
10,0 / 10
Cobertura de tareasImagen a video, Referencia a video, Texto a video
7,5 / 10
Personalización (LoRA)Sin compatibilidad con LoRA
0,0 / 5

Lo que MiniMax H3 X Turbo puede hacer en SpicyAPI

Tomado directamente de la especificación en vivo del modelo en SpicyAPI. Algunos creadores ofrecen más en sus propias apps; esto es lo que puedes llamar hoy desde nuestra API.

Capacidades por endpoint
FunciónImagen a videoReferencia a videoTexto a video
Resoluciones480p · 720p · 1080p480p · 720p · 1080p480p · 720p · 1080p
Duración del clip2–15 s2–15 s2–15 s
Relaciones de aspecto—16:9 · 9:16 · 1:1 · 4:3 · 3:4 · 21:916:9 · 9:16 · 1:1 · 4:3 · 3:4 · 21:9
Audio nativoSíSíSí
Entradas de referencia—14—
Control de semillaNoNoNo
Prompt negativoNoNoNo
Ver todos los parámetros y ejemplos de código →

Cuánto tarda de verdad

Medido con trabajos reales de los últimos 30 días, desde la solicitud hasta el resultado final. Los tiempos incluyen nuestras propias pruebas.

Tiempo mediano
63 s
Percentil 90
1,8 min
Espera por segundo de video
12,3 s por segundo de salida
Tasa de éxito
Recopilando datos

Un tiempo aparece cuando hay al menos 10 trabajos completados. Las tasas de éxito solo cuentan trabajos de clientes y siguen ocultas para todos los modelos hasta completar 30 días de tráfico.

Modelos que puedes considerar en lugar de MiniMax H3 X Turbo

¿Qué opinas?

¿Te resultó útil esta reseña?

Inicia sesión para votar.

Comentarios

Los comentarios se revisan de forma automática y por nuestro equipo. Eliminamos el spam, el contenido inapropiado, los ataques personales y cualquier cosa que nombre a una persona real sin su consentimiento. Si reportas un comentario, lo revisamos en un día hábil.

¿Probaste este modelo? Comparte lo que te funcionó. Necesitas una cuenta gratuita para comentar, así mantenemos fuera el spam.

Preguntas frecuentes sobre MiniMax H3 X Turbo

Prueba MiniMax H3 X Turbo por tu cuenta

Genera en Studio en pocos clics, o configura cada parámetro en el Playground y consulta el precio exacto antes de ejecutar.

0,12 US$ por clip de 5 segundos

Más reseñas

Ver las clasificaciones →
  • Una mujer con un vestido de seda esmeralda entre grandes hojas tropicales en un invernadero de cristal al atardecer

    Wan 3.0

    Wan 3.0 para tomas de 30 s baratas y sin censura, con sonido, desde texto, una imagen o referencias; no cuando la escena debe parar donde para tu guion.

    Spicy Index
    87,7
    Freedom Score
    96,0

    Probado el 24 sept 2026

    Leer la reseña
  • A woman laughing in a swimming pool at night, hair swept back, pink and teal neon light rippling across the water around her.

    MiniMax H3 HighSpeed

    Pick MiniMax H3 HighSpeed if you want 768p takes with sound that we saw come back as written at Very strong; for cheaper drafts or 2K, look at Lite or H3.

    Spicy Index
    52,0
    Freedom Score
    20,0

    Probado el 27 oct 2026

    Leer la reseña
  • Una mujer con un vestido de seda dorada camina entre las olas al atardecer y la tela ondea detrás de ella

    Seedance 2.5

    Seedance 2.5 cuando una toma de 30 s debe seguir el guion o un gran reparto de referencias; para clips sin censura baratos, MiniMax H3 cuesta mucho menos.

    Spicy Index
    82,2
    Freedom Score
    83,6

    Probado el 24 sept 2026

    Leer la reseña