When a model is listed as a verified model below, it means there is a contract in place with the provider that prohibits the provider from using customer content to train or improve its generative AI models when used via Figma Weave.
| Model Name | Delivered Through |
|---|---|
| Hunyuan 3D V2.0 | FAL |
| Hunyuan Video to Video | FAL |
| Ideogram V3 | FAL |
| Ideogram V3 Character | FAL |
| Ideogram Replace Background | FAL |
| Kling AI Avatar Pro | FAL |
| Kling 3 | FAL |
| Kling 3O Edit Video | FAL |
| Kling Element | FAL |
| Koloros Virtual Try-On | FAL |
| Omnihumanv1.5 | FAL |
| PixVerse Lipsync | FAL |
| Qwen Image Edit Plus | FAL |
| Qwen Image Edit 2511 | FAL |
| Qwen-3 TTS 1.7B Voiceover | FAL |
| Rodin | FAL |
| Rodin V2 | FAL |
| Seededit 3.0 | FAL |
| Seedream V4 Edit | FAL |
| Seedream V5 Edit | FAL |
| Seedream 4.5 | FAL |
| Seedance 1.5 Pro | FAL |
| Seedance 2.0 Fast | FAL |
| Seedance 2.0 Standard | FAL |
| Seedance 2.0 Reference Standard | FAL |
| Seedance 2.0 Reference Fast | FAL |
| Seedance 2.0 Mini | FAL |
| Seedance 2.5 | FAL |
| Seedance 2.5 Reference | FAL |
| Seed Audio 1.0 | FAL |
| Topaz Upscale | FAL |
| Topaz Video Upscaler | FAL |
| Trellis | FAL |
| Trellis 3D V2 | FAL |
| Veo 2 image to video | FAL |
| Veo 2 text to video | FAL |
| Veo 3 image to video | FAL |
| Veo 3 text to Video | FAL |
| Veo 3.1 Image to Video | FAL |
| Veo 3.1 text to Video | FAL |
| Wan 2.2 | FAL |
| Wan 2.5 | FAL |
| Wan 2.7 Video | FAL |
| Wan 2.7 Image | FAL |
| Wan VaceDepth | FAL |
| Wan VaceOutpaint | FAL |
| Wan VacePose | FAL |
| Wan VaceReframe | FAL |
| HeyGen Avatar 5 | FAL |
| HeyGen Avatar 5 Lip Sync | FAL |
| Gemini 3 - Nano Banana Pro | FAL / Google |
| Grok Edit Image | FAL |
| Grok Video Edit | FAL |
| Gemini 3.1 - Nano Banana 2 | FAL / Google |
| Gemini 2.0 and 2.5 (Nano Banana) Flash | |
| Gemini 3.1 Flash TTS | FAL / Google |
| Google 3.1 Omni Flash | FAL / Google |
| Google Omni | FAL / Google |
| Google Omni Flash 1.1 | FAL / Google |
| Google Chirp 3 | |
| Google Lyria 3 | |
| Google Lyria 3 Pro | |
| Chirp 3 Transcription | |
| NvidiaConsistory | Nvidia |
| Nvidia Cosmos 3 Super | FAL |
| Nvidia Cosmos 3 Super Video | FAL |
| Dalle 3 | OpenAI |
| GPT image 1 | OpenAI |
| GPT image 1 Edit | OpenAI |
| GPT image 1.5 Edit | OpenAI |
| GPT image 2 | OpenAI |
| GPT image 2 Edit | OpenAI |
| GPT Image 5 (Chat) | OpenAI |
| Sora 2 | OpenAI |
| Runway Act 2 | Runway |
| Runway Aleph | Runway |
| Runway Gen 3 Video | Runway |
| Runway Gen 4 Video | Runway |
| Runway Gen 4.5 Video | Runway |
| Runway Video | Runway |
| Runway Aleph 2.0 | Runway |
| Image To Image | Stability |
| SD3 Content-Aware Fill | Stability |
| SD3 Inpaint | Stability |
| SD3 Outpaint | Stability |
| SD3 RemoveBackground | Stability |
| Sketch to Image | Stability |
| Pixverse 4.5 | Replicate |
| Pixverse 6 | FAL |
| MiniMax H3 Video | FAL |
| MiniMax H3 Reference to Video | FAL |
| Minimax Video 01 | Replicate |
| Minimax Video Director | Replicate |
| Minimax Image 01 | Replicate |
| Nvidia Sana | Replicate |
| Recraft CrispUpscaler | Replicate |
| Recraft V3 Vector | Replicate |
| Recraft V3 | Replicate |
| Recraft V4 | FAL |
Last Updated: September 8, 2026
Use of unverified models is subject to the terms of your agreement with us as well as the terms and privacy policies of the applicable model provider. As a result, those model providers have additional rights to your data, which may include the right to train their generative AI models. You are solely responsible for reviewing and vetting unverified models (including all applicable terms and policies) for your use case. By using an unverified model, you agree to the applicable unverified model’s applicable terms and acknowledge their privacy policies. For your convenience, we have provided references to the provider’s terms and privacy policies below.
| --- | --- | --- | --- |
Last Updated: April 24, 2026