Genfeed.ai
Start creating

Models

The models available in Genfeed, read directly from the product registry.

Start creating
Explore Studio
A curated museum cabinet of distinct optical glass lenses and sculptural metallic creative tools, each chamber has its own restrained lighting and material
Live catalog
95Models
10Formats
7Providers
Read from the same public registry as the app and refreshed hourly.See how they compare on the benchmarkFind a model with the free AI model selector
Current availability

Choose by output.

Availability, defaults, and capabilities update from the registry. Nothing on this page is maintained by hand.
01

Video

17 models
  1. 01

    MiniMax H3

    Default
    MiniMax H3 — multimodal text, first/last-frame, and reference video generation with native audio at 768P or 2K.
    Replicate
  2. 02

    Gemini Omni Flash

    Google Gemini Omni Flash via fal — synchronized-audio video from text, a first image, or up to three reference images.
    Fal
  3. 03

    Seedance 2.5

    ByteDance Seedance 2.5 — flagship multimodal video with native audio (up to 30s). Expensive; prefer short drafts.
    Replicate
  4. 04

    H3 Max Director

    MiniMax H3 Max Director — realtime steered video billed per elapsed second, including idle time, with a 60-second session minimum.
    Fal
  5. 05

    Hailuo 2.3 Fast

    MiniMax Hailuo 2.3 Fast — image-to-video from a required first frame. Legacy; MiniMax H3 is the successor.
    Replicate
  6. 06

    Kling Avatar V2

    Kling Avatar V2 (video)
    Replicate
  7. 07

    Kling V1 6 Pro

    Kling V1 6 Pro (video)
    Replicate
  8. 08

    Kling V2 1

    Kling V2 1 (video)
    Replicate
  9. 09

    Kling V2 1 Master

    Kling V2 1 Master (video)
    Replicate
  10. 10

    Kling V2 5 Turbo Pro

    Kling V2 5 Turbo Pro (video)
    Replicate
  11. 11

    Kling V3 Omni Video

    Kling V3 Omni Video (video)
    Replicate
  12. 12

    Kling V3 Video

    Kling V3 Video (video)
    Replicate
  13. 13

    Veo 3

    Veo 3 (video)
    Replicate
  14. 14

    Veo 3 1

    Veo 3 1 (video)
    Replicate
  15. 15

    Veo 3 1 Fast

    Veo 3 1 Fast (video)
    Replicate
  16. 16

    Veo 3 Fast

    Veo 3 Fast (video)
    Replicate
  17. 17

    Wan 2.2 I2V Fast

    Wan 2.2 I2V Fast — image-to-video with optional interpolation at 480p or 720p.
    Replicate
02

Video Edit

3 models
  1. 01

    Lipsync 2

    sync lipsync-2 — re-times the mouth of an existing video to supplied audio. 17 credits per second, 85 for a 5 s clip.
    Replicate
  2. 02

    Lipsync 2 Pro

    sync lipsync-2-pro — higher-fidelity lip-sync for an existing video, including teeth and facial detail. 28 credits per second, 140 for a 5 s clip.
    Replicate
  3. 03

    Reframe Video

    Reframe Video (video-edit)
    Replicate
03

Video Upscale

2 models
  1. 01

    ByteDance Video Upscaler

    ByteDance vCube video upscaler — up to 4K/60fps with scene presets (aigc/ugc/film).
    Replicate
  2. 02

    Video Upscale

    Video Upscale (video-upscale)
    Replicate
04

Image

40 models
  1. 01

    FLUX.3

    FLUX.3 Image — generation and reference editing with up to ten images, five resolutions and one output.
    Replicate
  2. 02

    GPT Image 2.5 Flare

    OpenAI GPT Image 2.5 Flare — fastest 2.5 image model, high-quality everyday generation and editing.
    Replicate
  3. 03

    GPT Image 2.5 Sunburst

    OpenAI GPT Image 2.5 Sunburst — most capable 2.5 image model for precise edits and detailed control.
    Replicate
  4. 04

    Seedream 5 Pro

    ByteDance Seedream 5 Pro — flagship image (1K/2K, up to 10 reference images).
    Replicate
  5. 05

    Nano Banana 2 Lite

    Default
    Google Nano Banana 2 Lite — fast 1K image generation and editing with up to 14 references.
    Replicate
  6. 06

    Flux 1 1 Pro

    Flux 1 1 Pro (image)
    Replicate
  7. 07

    FLUX.1 Schnell

    FLUX.1 Schnell — legacy speed-focused image model for development and compatibility. Production Auto uses current recommended models.
    Replicate
  8. 08

    Flux 2 Dev

    Flux 2 Dev (image)
    Replicate
  9. 09

    Flux2 Dev

    Flux2 Dev (image)
    Genfeed Ai
  10. 10

    Flux2 Dev Pulid

    Flux2 Dev Pulid (image)
    Genfeed Ai
  11. 11

    Flux2 Dev Pulid Lora

    Flux2 Dev Pulid Lora (image)
    Genfeed Ai
  12. 12

    Flux2 Dev Pulid Upscale

    Flux2 Dev Pulid Upscale (image)
    Genfeed Ai
  13. 13

    Flux 2 Flex

    Flux 2 Flex (image)
    Replicate
  14. 14

    Flux2 Klein

    Flux2 Klein (image)
    Genfeed Ai
  15. 15

    FLUX.2 klein 4B

    FLUX.2 klein 4B — sub-second draft generation and reference editing with up to five images. Storyboard uses it for black-and-white sketch stills.
    Replicate
  16. 16

    Flux 2 Pro

    Flux 2 Pro (image)
    Replicate
  17. 17

    Flux Dev

    Flux Dev (image)
    Genfeed Ai
  18. 18

    Flux Dev Pulid

    Flux Dev Pulid (image)
    Genfeed Ai
  19. 19

    Flux Kontext Pro

    Flux Kontext Pro (image)
    Replicate
  20. 20

    Gen4 Image Turbo

    Gen4 Image Turbo (image)
    Replicate
  21. 21

    GPT Image 1.5

    OpenAI GPT Image 1.5 — previous GPT Image generation and editing model.
    Replicate
  22. 22

    GPT Image 2

    OpenAI GPT Image 2 — instruction-following image generation and editing.
    Replicate
  23. 23

    Ideogram Character

    Ideogram Character (image)
    Replicate
  24. 24

    Ideogram V3 Balanced

    Ideogram V3 Balanced (image)
    Replicate
  25. 25

    Ideogram V3 Quality

    Ideogram V3 Quality (image)
    Replicate
  26. 26

    Ideogram V3 Turbo

    Ideogram V3 Turbo (image)
    Replicate
  27. 27

    Imagen 3

    Imagen 3 (image)
    Replicate
  28. 28

    Imagen 3 Fast

    Imagen 3 Fast (image)
    Replicate
  29. 29

    Imagen 4

    Imagen 4 (image)
    Replicate
  30. 30

    Imagen 4 Fast

    Imagen 4 Fast (image)
    Replicate
  31. 31

    Imagen 4 Ultra

    Imagen 4 Ultra (image)
    Replicate
  32. 32

    Nano Banana

    Replicate Nano Banana image generation model
    Replicate
  33. 33

    Nano Banana 2

    Replicate Nano Banana 2 image generation model
    Replicate
  34. 34

    Nano Banana 2.1

    Google Nano Banana 2.1 — image generation and editing at 1K, 2K, and 4K with up to 14 references. The seeded USD is the 1K output price; 2K and 4K bill from the reviewed rate sheet.
    Replicate
  35. 35

    Nano Banana Pro

    Replicate Nano Banana Pro image generation model
    Replicate
  36. 36

    Qwen Image

    Qwen Image (image)
    Replicate
  37. 37

    Seedream 4

    Seedream 4 (image)
    Replicate
  38. 38

    Seedream 4 5

    Seedream 4 5 (image)
    Replicate
  39. 39

    Z Image Turbo

    Z Image Turbo (image)
    Genfeed Ai
  40. 40

    Z Image Turbo Lora

    Z Image Turbo Lora (image)
    Genfeed Ai
05

Image Edit

3 models
  1. 01

    FLUX.3 Edit

    FLUX.3 Image — generation and reference editing with up to ten images, five resolutions and one output.
    Replicate
  2. 02

    Ideogram 4.5

    Default
    Ideogram 4.5 — instruction-based image editing with up to five source images and an optional mask. Medium quality preserves source proportions by default.
    Replicate
  3. 03

    Reframe Image

    Reframe Image (image-edit)
    Replicate
06

Image Upscale

1 model
  1. 01

    Image Upscale

    Image Upscale (image-upscale)
    Replicate
07

Voice

4 models
  1. 01

    ElevenLabs v4

    Default
    ElevenLabs v4 — the most expressive ElevenLabs voice model, 90+ languages, up to 10,000 characters. 27 credits per 1,000 characters.
    Elevenlabs
  2. 02

    ElevenLabs Flash v2.5

    ElevenLabs Flash v2.5 — fastest and cheapest ElevenLabs voice, 32 languages, up to 40,000 characters. 14 credits per 1,000 characters.
    Elevenlabs
  3. 03

    ElevenLabs Multilingual v2

    ElevenLabs Multilingual v2 — stable long-form narration in 29 languages, up to 10,000 characters. 27 credits per 1,000 characters.
    Elevenlabs
  4. 04

    HeyGen Voice

    HeyGen Voice — speaks text of up to 5,000 characters in a HeyGen Voice clone your workspace owns. 10 credits per 1,000 characters.
    Heygen
08

Avatar

3 models
  1. 01

    HeyGen Avatar IV

    Default
    HeyGen Avatar IV — talking-avatar video from a trained HeyGen look. 65 credits for a 5 s photo-avatar clip, 135 for digital twins and studio avatars.
    Heygen
  2. 02

    HeyGen Avatar V

    HeyGen Avatar V — the highest-fidelity engine for digital-twin looks. 200 credits for a 5 s clip.
    Heygen
  3. 03

    HeyGen Photo Avatar

    HeyGen Photo Avatar — animates a single photo into a talking video with Avatar IV. 13 credits per second, 65 for a 5 s clip.
    Heygen
09

Music

3 models
  1. 01

    Lyria 3 Pro

    Default
    Google Lyria 3 Pro via fal — full songs with vocals, lyrics, and multi-language support, up to 90s.
    Fal
  2. 02

    Eleven Music

    ElevenLabs Music via fal — full compositions with vocals and lyrics, 10s-90s.
    Fal
  3. 03

    Mureka V9

    Mureka V9 — lyrics-first song generation via a direct API integration, up to 90s.
    Mureka
10

Text

19 models
  1. 01

    Gemini 2.5 Flash Lite

    Default daily driver via OpenRouter — fast Flash-Lite, cheap tool turns
    Agent Chat
    Openrouter
  2. 02

    Nemotron 3 Ultra 550b A55b:Free

    Retired — superseded by DeepSeek V4 Flash
    Agent Chat
    Openrouter
  3. 03

    DeepSeek V4 Flash

    Default
    Fast, low-cost agent chat with current tool-calling support and 1M context
    Agent Chat
    Openrouter
  4. 04

    Claude Opus 5

    Most capable — reserve for hard planning and review
    Agent Chat
    Openrouter
  5. 05

    Claude Sonnet 5

    Balanced intelligence for long multi-step work
    Agent Chat
    Openrouter
  6. 06

    Deepseek R1

    Deepseek R1 (text)
    Replicate
  7. 07

    Gemini 2 5 Flash

    Gemini 2 5 Flash (text)
    Replicate
  8. 08

    Gemini 3.5 Flash Lite

    Stronger Flash Lite for light agentic work
    Agent Chat
    Openrouter
  9. 09

    Gemini 3.6 Flash

    Fast agentic reasoning with a large context window
    Agent Chat
    Openrouter
  10. 10

    Gemini 3 Pro

    Gemini 3 Pro (text)
    Replicate
  11. 11

    Gpt 5 2

    Gpt 5 2 (text)
    Replicate
  12. 12

    GPT-5.6 Luna

    Fast and cheap — short chat turns and quick edits
    Agent Chat
    Openrouter
  13. 13

    GPT-5.6 Sol

    Deepest reasoning — expensive, use deliberately
    Agent Chat
    Openrouter
  14. 14

    GPT-5.6 Terra

    Balanced for heavier agentic work
    Agent Chat
    Openrouter
  15. 15

    Grok 4 1 Fast

    Grok 4 1 Fast (text)
    Replicate
  16. 16

    Grok 4.6

    Real-time knowledge, fast responses
    Agent Chat
    Openrouter
  17. 17

    Kimi K3

    Agentic reasoning and multimodal work
    Agent Chat
    Openrouter
  18. 18

    Meta Llama 3 1 405b Instruct

    Meta Llama 3 1 405b Instruct (text)
    Replicate
  19. 19

    OpenRouter Free (Experimental)

    Experimental free router with capability-aware DeepSeek fallback
    Agent Chat
    Openrouter