Multimodal AI guide — text, image, video and audio under one OpenAI-compatible API

Most LLM aggregators stop at text. The moment your product needs to generate an image, edit a photo, animate a still into video, voice an ad, or compose music — you're back to managing three more vendors, three more bills, three more SDKs. Kunavo gives you all of it through the same OpenAI-compatible API. This guide covers when to reach for which modality, the model menu per modality, and how to chain them into production multimodal pipelines.

The modality menu (and what each is for)

• Text generation — Claude Opus/Sonnet/Haiku and GPT. Use for reasoning, chat, structured output. /docs/chat
• Image generation (text-to-image) — Nano Banana Pro, GPT-Image-2, Nano Banana 2. Picks vary by use case (see shootout below)
• Image editing (image-to-image) — Nano Banana Edit, GPT-Image-2 Edit. Replace text on packaging, swap backgrounds, restyle
• Video generation — Veo 3 Lite/Fast/Quality, Seedance 2 and Seedance 2 Fast, Wan 2.7. Text-to-video and image-to-video both supported
• Audio — Music — Suno V5 / V5.5. Full songs from a prompt

Browse the full catalog at /models.

Picking an image model — the 30-second rule

• Need clean text on the image (UI mockups, posters, signs)? → Nano Banana Pro. No contest
• Literal, detailed prompts? → GPT-Image-2. Strong prompt understanding; flat price at every resolution
• High volume where cost matters? → Nano Banana 2. Cheapest at 1K, fidelity holds up
• Cinematic with depth? → Nano Banana Pro. Bokeh is genuinely cinematic

Full deep-dive in our image models shootout — our live image-model comparison.

Picking a video model

Three families are live, and they bill on different units — Veo 3 per video, Seedance and Wan per second of output. Prices below are the catalog rates:

Because the endpoint is model-agnostic, switching families is a change to the model field and nothing else. Video quickstart in our Veo 3 video quickstart; per-model detail in the Seedance API and Wan API guides.

Generating an image — the basics

# Text-to-image: same OpenAI SDK, different model slug
img = client.images.generate(
    model="nano-banana-pro",          # Google's flagship — best text rendering
    prompt="a marketing hero for a fintech app, isometric, soft blue palette",
    size="1024x1024",
)
print(img.data[0].url)  # permanent: https://files.kunavo.com/...

Result URLs are permanent, for images as for videos: Kunavo archives every result to its files.kunavo.com CDN and hands back the archived file's URL.

Image-to-video — taking your existing assets to motion

You already have product photography on your site. Veo 3 can animate any of it into an 8-second product clip:

# Image-to-video: upload a hero image, animate into an 8-second product clip
# (every Veo clip is 8 seconds; the Veo models ignore "duration").
# The OpenAI SDK has no method for the video route, so post to it with the
# same client.
video = client.post(
    "/video/generations",
    body={
        "model": "veo-3",
        "prompt": "camera slowly orbits, soft cinematic light, product hero shot",
        "image_url": hero_image_url,
        "aspect_ratio": "9:16",   # vertical for Reels / TikTok / IG short
        "resolution": "1080p",
    },
    cast_to=object,
    options={"timeout": 600.0},   # a clip takes minutes
)
print(video["data"][0]["url"])

Marketing teams that previously spent $500-1500 per video shoot compress that to a clip from $0.18+ (Veo 3.1 Lite) to $1.60+ (Veo 3.1 Quality) plus about 30 seconds of compute. Works especially well for e-commerce hero animations, App Store / Play Store screenshots, social media product showcases.

Chaining modalities — the real superpower

The interesting workflows mix modalities in sequence. A marketing clip production line:

# Pipeline: GPT-Image generates a marketing hero, Veo 3 animates it,
# Suno scores the BGM. One API, one bill.
# Here client is AsyncOpenAI(api_key=..., base_url="https://api.kunavo.com/v1").
async def make_marketing_clip(product: dict) -> dict:
    # 1. Hero image
    img = await client.images.generate(
        model="gpt-image-2",
        prompt=f"hero shot of {product['name']}, premium aesthetic",
        size="1024x1792",
    )

    # 2. Animate to 8s vertical (the length of every Veo clip). The SDK has
    #    no method for the video and music routes, so post to them directly.
    video = await client.post(
        "/video/generations",
        body={
            "model": "veo-3",
            "prompt": f"camera orbits {product['name']}, cinematic lighting",
            "image_url": img.data[0].url,
            "aspect_ratio": "9:16",
        },
        cast_to=object,
        options={"timeout": 600.0},
    )

    # 3. BGM
    bgm = await client.post(
        "/audio/music",
        body={
            "model": "suno-v5",
            "prompt": "upbeat synth, 30 seconds, brand-friendly",
        },
        cast_to=object,
        options={"timeout": 600.0},
    )

    return {"video": video["data"][0]["url"], "bgm": bgm["data"][0]["url"]}

Run this for 100 SKUs in parallel with asyncio.gather() and you have a marketing catalog production pipeline. Total cost ~$1-2 per SKU end-to-end. Compare to a creative agency's $500-2000 per asset.

Practical considerations

• Result URL lifetimes: images and videos are both permanent on Kunavo's files.kunavo.com; a URL on any other host is the upstream's temporary one, returned when archiving failed (re-host that one)
• Generation time: image ~5-30s, video ~30s-3min, music ~30s. Use force-dynamic server-side, surface progress to users
• Failed generations are free — Kunavo never bills 4xx/5xx, so iterate freely on prompts
• Text rendering in video: Veo 3 struggles with multi-word text in video. Render text in post (CapCut, Final Cut, browser-side canvas) instead of asking the model to render it
• Real faces / IP: all major models refuse celebrity, political, and copyrighted character generation. Stick to original characters or licensed IP
• Aspect ratios: 9:16 (vertical short), 16:9 (landscape), 1:1 (feed), 4:5 (Instagram portrait). Pass via aspect_ratio on video; on images, use size="1024x1792" etc.

Common multimodal use cases (with starting prompts)

• E-commerce product hero animations — image-to-video, orbital camera, 8 seconds, 9:16
• Real estate virtual staging — image-edit to add furniture to empty rooms
• Marketing carousel / Reels production — batch image generation in your brand style + Veo animation
• Educational explainer clips — generate slides as images, animate them with Veo 3, score with Suno
• Podcast intros & audio branding — Suno music beds and stings for your show, generated from a prompt
• Game / app asset generation — character portraits, item icons, environment backgrounds via Nano Banana Pro / GPT-Image-2

When NOT to use multimodal

• Strict brand assets — model output drifts, even with tight prompts. Use AI for variations on a human-designed master, not to replace the master
• Faces of real people — refused by all major models; legal/IP risk if you bypass
• Live video / real-time generation — current models take 30s-3min per clip. Not yet suitable for live use cases
• High-stakes accuracy — medical imaging, legal evidence, scientific publication. Multimodal generation is creative, not factual

Multimodal under one bill is what Kunavo is really for. Start at /docs/images or /docs/video for endpoint reference, or browse /models filtered by category.

For the per-model price tables behind these pipelines, see the image generation API comparison and the video generation API comparison.

Frequently asked questions