NNyquest docs

Media Studio — Images, Voice, Music & Video

Nyquest's Studio panels turn a text prompt into media. There are four — Image Studio, Audio Studio (text-to-speech), Music Studio and Video Studio. Open the circular navigation button at the right of the header, then choose Studios and the studio you want. The studios are powered by Nyquest's direct model access, so you get first-party quality without wiring up your own provider keys.

The Studios are a Pro feature. Pro starts with your first wallet funding, and each generation draws from your wallet balance. On the free tier, each Studio shows a Go Pro notice instead of generating. Using your own provider key (BYOK) in Image Studio or Audio Studio works on any tier.

Image Studio

Choose an available image model from the picker, enter a prompt, and generate. Supported aspect ratios depend on the model. Images are saved in your gallery and can be downloaded or deleted. Google retired its direct Imagen 4 models; use an available image model instead. Your existing images remain accessible.

Audio Studio (text-to-speech)

Turn text into natural speech. In addition to the standard voices, Nyquest now offers 30+ native Google Gemini voices (Kore, Puck, Charon, Aoede, and more). Choose a Gemini TTS model, pick a voice, paste your text, and generate a downloadable audio file. You can add a short style instruction (e.g. "say it warmly") to shape delivery.

ElevenLabs voices are available too: ElevenLabs v4 (most expressive), v4 Turbo (faster) and Flash v2.5 (fastest), each with 21 premade voices such as Sarah, George and Alice. They are priced per character of text: $0.04 per 1,000 characters on v4 and $0.02 per 1,000 on v4 Turbo and Flash v2.5, about $0.02 to $0.04 per minute of speech. The Studio shows the estimate as you type, and the request is refused before generating if your balance can't cover it. ElevenLabs models don't use style instructions; on v4, put tags such as [whispering] or [laughs] in the text instead (tags count as characters).

The voice list follows the selected model. Use the play button beside a voice to preview a short sample before generating your full passage. A preview is a generation request and can incur provider charges; it follows the same account and BYOK requirements as speech generation. A failed preview shows an error beside the controls so you can try another available voice or model.

Music Studio

Generate original instrumental music from a text prompt using Google Lyria. Describe a mood, instrument, tempo, or genre — e.g. "a warm upbeat acoustic guitar melody, 90 bpm" — and Lyria composes a short instrumental clip (~30 seconds) you can play inline and download. Two tiers: Lyria 3 (Clip) and the higher-fidelity Lyria 3 Pro.

Music availability depends on the live service. If the panel reports that music generation is temporarily unavailable, generation is disabled. Check again later.

Video Studio

Generate short video clips from a text prompt using Google Veo 3.1. Video is different from the other three because it takes longer (roughly 1–2 minutes), so it runs as a background job:

  1. You submit a prompt and pick the aspect ratio and clip length (4, 6, or 8 seconds; Veo's max is 8s). The estimated cost is held from your balance up front.
  2. The clip renders in the background — the panel shows Rendering… and updates automatically when it's ready.
  3. When done, the video plays inline and can be downloaded.

Because video is the most expensive media type (a clip can cost a few dollars), two safeguards apply: you must have enough balance to cover the clip before it submits, and confirmed submission rejection or render failure releases the hold. An unknown provider outcome keeps the hold pending reconciliation; check the job before retrying. There's also a small cap on how many videos you can render at once.

Billing at a glance

Managed Studio media is billed from your wallet at provider cost. Image and Audio Studio BYOK requests use your provider account instead. OpenRouter media uses the provider-reported charge; direct Google models use their published rates and billable units. There is no Nyquest media surcharge. A missing provider charge shows Cost pending until it can be reconciled; no estimated debit is applied. A reported zero remains zero. Each completed generation appears in your Usage page and billing ledger (/billing/ledger). Nothing is charged if a request is rejected before it runs; video specifically refunds its hold if the render fails.

Using it from the API

Studio media generation is also available to API clients. See Developer API → Endpoints Reference for the current routes, request shapes, and the async job flow that video uses.