-
black-forest-labs/FLUX.1-dev
Text-to-Image โข 12B โข Updated โข 631k โข โข 14.2k -
black-forest-labs/FLUX.1-Kontext-dev
Image-to-Image โข 12B โข Updated โข 241k โข โข 2.78k -
Qwen/Qwen-Image
Text-to-Image โข 20B โข Updated โข 245k โข โข 2.58k -
Qwen/Qwen-Image-Edit
Image-to-Image โข 20B โข Updated โข 123k โข โข 2.49k
Inference Provider
VERIFIED
36,123 monthly requests
AI & ML interests
None defined yet.
Recent Activity
View all activity
Team members 3 private
The upstream repositories whose weights actually run behind WaveSpeed endpoints, if you would rather run them yourself.
Instruction-driven editing, outpainting and upscaling: give it an image and a sentence describing the change.
FP8, INT8, NF4 and precision-cast variants of open models. Each card says exactly which tensors changed and which were left alone.
Text-to-video, image-to-video and audio-driven avatar Spaces: Wan, Veo, Sora, Seedance, Hailuo and Vidu.
-
Wan 2.2
๐18Reference for Wan 2.2, Alibaba's open video models
-
Wan2.2-S2V
๐ฃReference for Wan2.2-S2V, audio-driven video generation
-
Wan 2.1 Ultra Fast (image to video)
๐ฌ40Wan 2.1 image-to-video at 480p, tuned for speed.
-
Wan 2.1 VACE 14B
๐ฌ10VACE: all-in-one video creation and editing.
Bring-your-own-key Spaces for hosted text-to-image models, plus reference pages for the open-weight ones. No key is stored.
The upstream repositories whose weights actually run behind WaveSpeed endpoints, if you would rather run them yourself.
-
black-forest-labs/FLUX.1-dev
Text-to-Image โข 12B โข Updated โข 631k โข โข 14.2k -
black-forest-labs/FLUX.1-Kontext-dev
Image-to-Image โข 12B โข Updated โข 241k โข โข 2.78k -
Qwen/Qwen-Image
Text-to-Image โข 20B โข Updated โข 245k โข โข 2.58k -
Qwen/Qwen-Image-Edit
Image-to-Image โข 20B โข Updated โข 123k โข โข 2.49k
Text-to-video, image-to-video and audio-driven avatar Spaces: Wan, Veo, Sora, Seedance, Hailuo and Vidu.
-
Wan 2.2
๐18Reference for Wan 2.2, Alibaba's open video models
-
Wan2.2-S2V
๐ฃReference for Wan2.2-S2V, audio-driven video generation
-
Wan 2.1 Ultra Fast (image to video)
๐ฌ40Wan 2.1 image-to-video at 480p, tuned for speed.
-
Wan 2.1 VACE 14B
๐ฌ10VACE: all-in-one video creation and editing.
Instruction-driven editing, outpainting and upscaling: give it an image and a sentence describing the change.
Bring-your-own-key Spaces for hosted text-to-image models, plus reference pages for the open-weight ones. No key is stored.
FP8, INT8, NF4 and precision-cast variants of open models. Each card says exactly which tensors changed and which were left alone.