Model reference · open weights
GAE-D64 is an open-weight video model from TencentARC. GAE-D64-1B (BF16) weighs 5.6 GB; the smallest configuration that runs it is RTX 3060 12 GB.
For self-hosting, compare this build's requirements below with the GPU you plan to rent.
Summary of the TencentARC/GAE-D64-1B model card, 2026-10-01
What it is
| Released by | TencentARC |
|---|---|
| Released | 2026-09-17 |
| VRAM | 5.6 GB for the weights |
What it runs on
| Card | Weights | Memory |
|---|---|---|
| RTX 3060 12 GB | fits | 11.6 GB |
| RTX 4060 Ti 16 GB | fits | 15.4 GB |
| RTX 3090 24 GB | fits | 23.4 GB |
| RTX 4090 24 GB | fits | 23.4 GB |
| RTX 5090 32 GB | fits | 31.0 GB |
| L40S 48 GB | fits | 44.0 GB |
| A100 80 GB | fits | 78.2 GB |
| H100 80 GB | fits | 78.1 GB |
| RTX PRO 6000 Blackwell 96 GB | fits | 93.8 GB |
| DGX Spark (GB10) 128 GB unified | fits | 107 GB |
| H200 141 GB | fits | 138 GB |
| B200 180 GB | fits | 176 GB |
From the model card
Weights for GAE-64: a ~1B-parameter temporal DiT
(hidden_size=[768, 2048], depth=[28, 6]) over a 64-channel
geometry-native latent, at 672×378 and V=81, with metric Plücker
camera conditioning and Qwen3-0.6B text.
This repo is self-contained for GAE-64 inference: codec, per-channel latent
statistics, DA3 feature-level stats, and the flow transformer. The frozen
DA3-GIANT encoder / DPT head comes from
depth-anything/DA3-GIANT-1.1.
From the GAE code repository:
scripts/run_demo.sh
Or, after the files are in ckpts/:
python scripts/generate.py \
--image examples/scenes/forest_lake_trail.jpg \
--prompt-file examples/scenes/forest_lake_trail.txt \
--flow-ckpt ckpts/flow_gae64.pt \
--codec-ckpt ckpts/gae_64.pt \
--output results/forest
Image-to-video uses 50 Euler steps and CFG 2 with one reference view.
Layout:
codec/ — safetensors + config.jsongae_64.pt — drop-in checkpoint for gae.load_codeclatent_stats_gae_64.pt — per-channel latent mean/stdda3_stats_giant_5ds.tar — DA3 feature-level normalization statstransformer/ — flow safetensors + config.jsonflow_gae64.pt — drop-in checkpoint for gae.load_flowQuoted from the model card on Hugging Face. The full card is behind the Hugging Face link above.
Running it yourself
Rent a machine by the hour. ComfyUI is installed on it. Open ComfyUI through the tunnel and load the workflow from the model's card on Hugging Face; choose this model's file in its loader.
# on your rented machine (the ssh line is on its page in the console)
# get REPO FILE FOLDER: one file into /workspace/models/FOLDER, where ComfyUI loads it from
get() { hf download "$1" "$2" --local-dir /workspace/hf-files && mkdir -p "/workspace/models/$3" && mv "/workspace/hf-files/$2" "/workspace/models/$3/$4"; }
# the model (3.5 GB)
get TencentARC/GAE-D64-1B transformer/diffusion_pytorch_model.safetensors diffusion_models
start-comfyui# on your computer, in a second terminal: ComfyUI in your browser at http://localhost:8188
# HOST, PORT and BOOKING are your machine's, from its page in the console
ssh -o HostKeyAlias=axforge-BOOKING -o StrictHostKeyChecking=accept-new -L 8188:localhost:8188 dev@HOST -p PORT