Instructions to use Wan-AI/Wan2.1-T2V-1.3B-Diffusers with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use Wan-AI/Wan2.1-T2V-1.3B-Diffusers with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("Wan-AI/Wan2.1-T2V-1.3B-Diffusers", dtype=torch.bfloat16, device_map="cuda") prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" image = pipe(prompt).images[0] - Notebooks
- Google Colab
- Kaggle
Wan2.1 T2V 1.3B: proposal for a community Apple Silicon video-API example
Proposing a Wan2.1 T2V-1.3B-specific local API example for Apple Silicon clients.
Rapid-MLX's catalog points this alias at the official Diffusers repository. Its video integration has explicit frame-count, checkpoint-area and request-shape checks; callers use asynchronous /v1/videos jobs rather than chat completions.
The proposed example should select this 2.1 checkpoint explicitly, document its supported text-to-video controls and retain a generated-video receipt. The existing general Rapid guide also discusses Wan2.2 TI2V; that guide is not qualification evidence for Wan2.1 and image conditioning should not be copied across versions. Model revision, resolved backend, dimensions/frame count and output verification would be acceptance criteria for a ready-to-run community recipe.
Concrete catalog entry: wan2.1-t2v-1.3b-bf16 β Wan-AI/Wan2.1-T2V-1.3B-Diffusers@0fad780a534b6463e45facd96134c9f345acfa5b. This is a publisher-owned artifact.
Source references (fixed Rapid commit): catalog, video guide, Wan contracts, Diffusers path tests.
Verification status: this proposal is based on reading the catalog, implementation and the linked documentation/tests. No fresh model inference or test-suite run was performed for this submission; source tests are not an end-to-end qualification. It requests a community documentation scope, not an official endorsement or a performance listing.
Disclosure: I maintain Rapid-MLX. Prepared with AI assistance under my direction. Anonymous usage-metadata telemetry is enabled by default; DO_NOT_TRACK=1 disables it. The original model and conversion terms continue to apply.