YepAPI

Command Palette

Search for a command to run...

NEW APIJuly 31, 2026·New API

48 new image & video models — GPT Image, FLUX.2, Recraft, Seedream 4.5, Kling 3.0, Hailuo 3 and more

34 image models and 14 video models — including real SVG vector output from Recraft — all on the same /v1/media/queue endpoint you already use.

We have just added 48 new image and video generation models to YepAPI — the biggest media expansion we have shipped. Image generation now includes OpenAI's [GPT Image 2](/ai-api/openai/gpt-image-2), the full [FLUX.2](/ai-api/black-forest-labs/flux.2-pro) family from Black Forest Labs, eleven [Recraft](/ai-api/recraft/recraft-v4.1) tiers with true SVG vector output, ByteDance's [Seedream 4.5](/ai-api/bytedance-seed/seedream-4.5), [Krea 2](/ai-api/krea/krea-2-large), Microsoft's [MAI Image](/ai-api/microsoft/mai-image-2.5-pro), Sourceful's [Riverflow](/ai-api/sourceful/riverflow-v2.5-pro), and [Grok Imagine](/ai-api/x-ai/grok-imagine-image-quality). Video adds [Kling 3.0](/ai-api/kwaivgi/kling-v3.0-pro), MiniMax [Hailuo 3](/ai-api/minimax/hailuo-3), [Runway Gen-4.5](/ai-api/runway/gen-4.5) and Aleph 2, Alibaba's [Wan](/ai-api/alibaba/wan-2.7) and HappyHorse, Grok Imagine Video, and [Seedance 1.5 Pro](/ai-api/bytedance/seedance-1-5-pro). Every one of them is the same call you already make: `POST /v1/media/queue`, then poll `GET /v1/media/status/{jobId}` — same `yep_sk_` key, same `{ ok, data }` envelope, no new integration.

What's new

  • GPT Image 2, GPT Image 1 & Mini, GPT-5 Image & Mini, GPT-5.4 Image 2 — OpenAI's image lineup from $0.02/image
  • FLUX.2 Pro, Max, Flex & Klein — Black Forest Labs' full family from $0.03/image
  • Recraft V4.1, V4 and V3 across 11 tiers — including Vector tiers that return real, editable SVG
  • Seedream 4.5 (ByteDance) with 18 aspect ratios, Krea 2 Large/Medium/Turbo, and Microsoft MAI Image 2.5 & 2.5 Pro
  • Riverflow 2.5 & 2 (Sourceful), Gemini 2.5 Flash Image, Gemini 3.1 Flash Lite Image, and Grok Imagine Image
  • Kling 3.0 Pro & Standard plus Kling O1 — 3–15 second clips with synced audio and first/last-frame control
  • Hailuo 3 at 2K with audio, Runway Gen-4.5 & Aleph 2, Wan 2.7 & 2.6, HappyHorse 1.1 & 1.0, Grok Imagine Video 1.5, and Seedance 1.5 Pro
  • Video billed per second from $0.08/sec (Wan 2.6) — no subscriptions, no generation credits
  • All on the unified `/v1/media/queue` + `/v1/media/status/{jobId}` flow, with short aliases like `flux`, `gpt-image`, `recraft`, `kling`, `hailuo`, `wan` and `seedream`

What's new

34 image models and 14 video models landed at once. On the image side you get the current flagships — GPT Image 2 for prompt fidelity and in-image text, FLUX.2 Pro and Max for photorealism, Seedream 4.5 for the widest aspect-ratio range on the platform — plus specialists: Recraft for design work and true SVG vector output, Krea 2 for art-directed styling, MAI Image for Microsoft's own generator, and Riverflow for packaging and product visuals. On the video side, Kling 3.0 brings believable motion with synced audio and first/last-frame control, Hailuo 3 renders at 2K with audio, Runway Gen-4.5 and Aleph 2 bring Runway's cinematic language without a monthly plan, and Wan, HappyHorse, Grok Imagine Video, and Seedance 1.5 Pro round out the range from $0.08 to $0.59 per second.

Vector output you can actually edit

The Recraft Vector tiers return real SVG — paths, fills, and groups you can open in Figma, Illustrator, or Inkscape — not a raster image that looks like vector art. Set `options.outputFormat` to `svg` and you get a scalable design file back from a text prompt, which makes automated icon sets and logo exploration a single API call.

How to call them

Nothing changes in how you integrate: submit the job to `/v1/media/queue` with the model ID, then poll `/v1/media/status/{jobId}` until it reports `completed`. Options are per-model — aspect ratio, resolution, duration, output format, audio, and first/last frame images.

curl -X POST https://api.yepapi.com/v1/media/queue \
  -H "x-api-key: $YEPAPI_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "black-forest-labs/flux.2-pro",
    "prompt": "A weathered brass diving helmet on wet slate, cold window light",
    "options": { "aspectRatio": "16:9", "outputFormat": "png" }
  }'

Pricing

Images are billed per image, from $0.02 (GPT Image 1 Mini, GPT-5 Image Mini) up to $0.70 (Riverflow 2 Pro). Video is billed per second of generated footage, from $0.08/sec (Wan 2.6 at 480p) to $0.59/sec (Runway Aleph 2), with higher resolutions costing more on models that offer them. Everything comes out of the same prepaid balance as your text, SEO, and scraping calls — no subscriptions, no generation credits, no minimums.

Generate your first image or clip

Grab an API key and call the media queue in under a minute. $5 free credit on signup, no card required — enough for dozens of images.

Browse the media models