studiark.com

MODELS

20 models routed today. More post-launch.

Studiark routes prompts to the best-fit model per stage of the pipeline. One credit pool, one project, one rights record. Additional image and video providers may be added after policy and provider review; voice cloning remains unavailable.

AI VIDEO

Video generation

4 models · updated May 2026

Per-scene Veo renders from approved stills. Tier and quality vary by Veo variant — our pick is marked with a star.

  1. Veo 3.1 Lite

    Google Vertex AI

    Lowest-cost current video tier for fast exploration.

    NEW budgetlow-cost
  2. Veo 3.1 Fast

    Google Vertex AI

    Budget tier video — used on Free and Starter.

    NEW budgetfast
  3. Veo 3.1

    Google Vertex AI

    Standard cinematic photorealism on Pro.

    NEW photorealcinematic
  4. Veo 3.1 Audio

    Google Vertex AI

    Native audio + SFX generation on Business and Enterprise.

    NEW photorealnative audio

AI IMAGE

Image generation

5 models · updated May 2026

Stills first, then motion. Imagen, GPT Image, and Nano Banana cover budget through quality-leader tiers — same workspace, same rights record.

  1. GPT Image 1.5

    OpenAI

    Quality leader at MVP (LM Arena Elo 1,264). Pro and above.

    NEW quality leadertypography
  2. Imagen 4 Ultra

    Google Vertex AI

    Up to 4K, Pro and above.

    premium4K
  3. Imagen 4

    Google Vertex AI

    High-quality 1024px stills, Starter and above.

    photorealhigh quality
  4. Imagen 4 Fast

    Google Vertex AI

    Budget stills on Free.

    budget
  5. Nano Banana

    Google Gemini API

    Gemini 2.5 Flash Image — fast, conversational image editing.

    fastlow-cost

AI MUSIC & VOICE

Music and voice

5 models · updated May 2026

Lyria for music, OpenAI TTS for narration, and Whisper for transcription. Voice cloning is unavailable.

  1. Lyria 3 Pro

    Google Gemini API

    Full songs with verses, choruses, and bridges — Pro and above.

    NEW musicflagship
  2. Lyria 3 Clip

    Google Gemini API

    30-second beds and trailer cues — Starter and above.

    fastmusic
  3. OpenAI TTS

    OpenAI

    8 built-in voices, multilingual narration — Starter and above.

    fastlow-cost
  4. OpenAI TTS HD

    OpenAI

    Higher-fidelity narration on Pro and above.

    narration
  5. Whisper

    OpenAI

    Transcription for captions and voice-to-prompt.

    utility

PLANNING & ORCHESTRATION

Planning models

6 models · updated May 2026

Reasoning models behind Studiark for iOS: parsing briefs into multi-scene plans, prompt assistance, and metadata.

  1. GPT-5.5

    OpenAI

    Default planner for Studiark iOS briefs.

    NEW planningflagship
  2. GPT-5.5 Pro

    OpenAI

    Long-context planning for multi-scene films.

    planning
  3. Claude Sonnet 4.6

    Anthropic

    Moderation and orchestration fallback.

    moderation
  4. Claude Haiku 4.5

    Anthropic

    Budget moderation and first-pass checks.

    moderationlow-cost
  5. Gemini 2.5 Flash

    Google

    Default text orchestration — best value at scale.

    planninglow-cost
  6. Gemini 2.5 Flash Lite

    Google

    Cheapest routing for first-pass checks.

    utilitylow-cost

EXPLORE STUDIARK

One subscription. Every shipped model.

See how Studiark for iOS routes the right model for each shot, or use the API documentation to plan a programmatic integration.