Models
Last updated April 5, 2026
Choose the right
model for the agent job.
AnyCap exposes multimodal models through one capability runtime and one CLI. This page helps teams choose the right model for a given agent workflow instead of treating every image or video request the same way.
Answer-first summary
The current public AnyCap model catalog includes image generation models for first-pass output and revision loops, video generation models for premium or production-friendly motion work, and a prompt-based music model for soundtrack drafts. The right choice usually depends on whether the job starts from a blank prompt or an existing asset, how much polish the first pass needs, and how much speed or cost efficiency matters in the workflow.
How to choose the right model
- Start with the output type: image, video, or music.
- Then decide whether the task needs a polished first pass, faster iteration, or revision from an existing asset.
- Use the model guide pages when the choice depends on motion style, editing workflow, or cost tradeoffs.
Visual guide
This illustration is a quick visual map of the current catalog: image models on one side, video models on another, and music generation as a separate capability lane inside the same agent runtime. It was generated with Nano Banana 2 to keep the page's visual language aligned with the model catalog itself.
Current model comparison
These are the current public models exposed through AnyCap. Credit ranges come from the same pricing inventory used on the pricing page, so the hub and pricing page stay aligned.
Image generation
Charged per call. Supports text-to-image and image-to-image modes.
| Model | Modes | Credits / call | Best for |
|---|---|---|---|
| FLUX.1 Kontext Max | text-to-image, image-to-image | varies | Design-heavy image generation and contextual edits where prompt adherence, visual richness, and iterative refinement matter. |
| GPT Image 2 | text-to-image, image-to-image | varies | General-purpose image generation and image edits when the workflow benefits from OpenAI's multimodal image model family. |
| Nano Banana Pro | text-to-image, image-to-image | ~7 | Targeted image editing and revision loops from an existing visual. |
| Nano Banana 2 | text-to-image, image-to-image | ~4 | Fast, scalable image generation and high-volume iteration. |
| Qwen Image | text-to-image, image-to-image | varies | Bilingual or instruction-heavy visual work, especially when an agent needs a model associated with the Qwen multimodal family. |
| Seedream 4.5 | text-to-image, image-to-image | varies | Everyday image generation, image transformation, and iterative editing where stable structure preservation matters. |
| Seedream 5 | text-to-image, image-to-image | ~2 | Polished first-pass image generation from a text prompt. |
Video generation
Charged per second of generated output. Supports text-to-video and image-to-video modes.
| Model | Modes | Credits / sec | Best for |
|---|---|---|---|
| Hailuo 2.3 | text-to-video, image-to-video | varies | Short narrative clips, expressive character motion, visual storytelling, and reference-image animation. |
| Veo 3.1 | text-to-video, image-to-video | ~20 | Premium text-to-video output when the first pass needs to look stronger. |
| Veo 3.1 Fast | text-to-video, image-to-video | varies | Rapid creative iteration and preview generation when an agent wants the Veo family with faster turnaround. |
| Sora 2 Pro | text-to-video, image-to-video | varies | High-end narrative, cinematic, product, and realistic video generation when teams want an OpenAI video model through the same CLI. |
| Seedance 1.5 Pro | text-to-video, image-to-video | ~14 | Steady production-friendly video workflows and repeatable image-to-video jobs. |
| Seedance 2.0 | text-to-video, image-to-video | varies | High-quality cinematic and product video workflows where agents need the newer Seedance model entry. |
| Seedance 2.0 Fast | text-to-video, image-to-video | varies | Previewing, ideation, and high-volume video iteration when an agent needs faster turnaround. |
| Kling 3.0 | text-to-video, image-to-video | ~9 | Cinematic motion and flexible image-to-video workflows. |
| Kling O1 | image-to-video | varies | Product demos, stylized motion design, and image-conditioned clips where the source frame should drive the video. |
Music generation
Charged per second of generated audio.
| Model | Modes | Credits / sec | Best for |
|---|---|---|---|
| ElevenLabs Music | text-to-music | ~1 | Prompt-based soundtrack drafts inside the same agent runtime. |
| Mureka V8 | text-to-music | varies | Songwriting, vocal-oriented drafts, and audio content production when an agent needs an alternative to Suno or ElevenLabs Music. |
| Suno V5 | text-to-music | varies | Structured songs, vocal demos, and full-track concepts that need lyrics, mood, and arrangement guidance. |
| Suno V5.5 | text-to-music | varies | Current Suno music generation workflows, complete track drafts, vocal concepts, and high-iteration song ideas. |
Image generation
Video generation
Music generation
FAQ
How do I choose between Seedream 5, Nano Banana Pro, and Nano Banana 2?
Use Seedream 5 when the workflow needs a stronger first-pass image from a prompt, Nano Banana Pro when the job starts from an existing image and needs revisions, and Nano Banana 2 when speed, throughput, or repeated iteration matters more.
How do I choose between Veo 3.1, Kling 3.0, and Seedance 1.5 Pro?
Use Veo 3.1 when the first video pass needs to look more premium from a text brief, Kling 3.0 when the workflow leans more on cinematic motion or flexible image-to-video work, and Seedance 1.5 Pro when the team wants a steadier production-oriented default.
Do all AnyCap models use the same CLI and auth flow?
Yes. AnyCap exposes these models through the same capability runtime, CLI, and auth flow, so teams do not need a separate provider integration path for each model page listed here.