Models

8 articles

• How do I choose a voice on a text-to-speech model?Select the TTS model in the Workspace and open the Voice dropdown in the parameters panel — ElevenLabs v3 offers 21 named voices with gender and accent shown, and the default is Bella (female, American); type the text in the prompt box and press Generate.
• How do resolution and duration change the price?Video models bill per second, and most multiply that rate by resolution — Seedance 2.0 charges 0.465× at 480p, 1× at 720p, 2.25× at 1080p and 5.13× at 4K — so a 10-second 1080p clip costs about four and a half times a 5-second 720p one; the Workspace estimate updates as you change either.
• What do Fast, Standard and Pro mean in a model name?They are the vendor's own quality tiers of the same model: Fast trades detail for speed and price, Pro is the highest quality at the highest price, and Standard sits between — for example a 5-second Veo 3.1 clip is 800 credits on Lite, 1,280 on Fast and 3,200 on the standard model.
• What does the New badge on a model mean?The amber New chip marks a model added to LUVI recently; it is set with an expiry date when the model is onboarded — usually about a month after release — and disappears by itself when that date passes.
• What is a prompting guide?A prompting guide is LUVI's written manual for one model family — the dialect it responds to, the words that break it and the parameters that matter — and it is applied automatically by the prompt-check feather in the Workspace and by Claude or ChatGPT through the connector.
• What is on a model page?Each page at /models/{family}/{model} shows the model's starting price in credits, its parameters with defaults and allowed values, the inputs it accepts, its prompting guide when one exists, and an Open in Workspace button that selects it for you.
• Which AI model should I use?Start from what you are making — a still, a clip with sound, a talking head, a voice, a track or a mesh — then pick the family whose strengths match; each family's page lists what it is best for and what it costs.
• Which models generate sound with the video?Seedance 2.0 and 2.5, Veo 3.1, Gemini Omni, MiniMax H3, HappyHorse, FLUX 3 Video, Grok Imagine Video and Kling 3 all produce the soundtrack in the same pass as the picture — dialogue, effects, ambience and music — but they differ in whether audio is on by default and whether it costs extra.