Models · Multi-view depth

VGGT-Omega

VGGT-Omega (Meta): the successor of VGGT with a DINOv3 backbone, cameras and depth from many images. Non-commercial licence.

stable Fork of facebookresearch/vggt-omega@48b23c8 · backends: portable, kernels

  • Paper VGGT-Ω CVPR 2026 · first released 14 May 2026

Checkpoint vggt-omega-1b-512

Paper 2026 Added in 0.1.0 Evaluated with 0.2.0

Kernels, RTX 5060 Ti 16 GB 467 ms 336x518, fp16
vs the faster original 1.9× original fp16: 906 ms
Peak memory 3,404 MiB original: 7,495 MiB
Accuracy vs original fp32 −0.0001 kernels AbsRel, DTU scans 1 (0.0050 original)

Use it

pip install fastforks

Lean install with only what this model needs: pip install "fastforks-core[nc]"

from fastforks import multiview

model = multiview.load("vggt-omega-1b-512")  # kernels on a supported GPU, else portable PyTorch
result = model(images)
result.depth  # also: confidence, intrinsics, cam_to_world
result.backend  # what ran: mode, device, dtype, kernels

The weights are non-commercial: fastforks downloads them only with FASTFORKS_NONCOMMERCIAL=1 set.

fastforks bench vggt-omega-1b-512

Times the model on this machine. load(…, mode="portable") or mode="kernels" picks a backend; usage guide.

Speed and memory

Every backend on one GPU (RTX 5060 Ti 16 GB) and the same input, the authors’ code included. Kernels runs fp16. Portable runs fp16 on CUDA.

Input: synthetic, 336x518

BackenddtypeTime per call (ms)vs original fp16Peak memory
allocated / reserved
Load (s)First call (s)
Original fp32float32 (the authors' autocast off)fp32 26500.34×5,765 MiB/5,912 MiB12.32.64
Original fp16original default VGGTOmega(autocast=True): bfloat16 aggregator, float32 headsfp16 906–7,495 MiB/7,654 MiB16.31.10
Portable fp32fp32not measured ––––
Portablefp16 7261.2×3,421 MiB/3,518 MiB3.30.73
Kernelsfp16 4671.9×3,404 MiB/3,494 MiB5.20.75
Portable + torch.compile 5891.5×3,082 MiB/3,154 MiB66 (build)–
TensorRT fp16fp16 3772.4×2,994 MiBenginenot comparable187 (build)–

Measured on RTX 5060 Ti 16 GB, torch 2.14.1+cu130, 2026-11-03 to 2026-11-16. First call, 3 warm-up, peak reset, median of 3 rounds x median of 5. Ratios divide two times of this table (same GPU, same input). Original fp16 is the authors’ own precision option, else torch.autocast. torch.compile and TensorRT: build time instead of load time. TensorRT memory (engine, not comparable) is the GPU memory the engine took, read from the driver because TensorRT allocates outside PyTorch, so it can't be compared with the PyTorch peaks of the other rows.

Other devices

DeviceBackenddtypems / callPeak memoryOutput vs referenceInput
Apple M1 Pro 16 GB MPS, torch 2.14.1, 2026-11-12Originalfp327828 9,079 MiB processmatches336x518
Portablefp167242 8,602 MiB processmatches336x518
Portablefp327938 8,221 MiB processmatches336x518
Kernelsfp16not implemented–336x518
NVIDIA L4 CUDA, torch 2.14.1+cu130, 2026-11-07Portablefp16758 3,380 MiB/3,476 MiB–336x518
Kernelsfp16599 3,380 MiB/3,494 MiB–336x518
NVIDIA L40S CUDA, torch 2.14.1+cu130, 2026-11-07Portablefp16199 3,380 MiB/3,476 MiB–336x518
Kernelsfp16166 3,380 MiB/3,494 MiB–336x518
NVIDIA A10 CUDA, torch 2.14.1+cu130, 2026-11-07Portablefp16578 3,380 MiB/3,476 MiB–336x518
Kernelsfp16447 3,380 MiB/3,434 MiB–336x518
NVIDIA A100-SXM4-40GB CUDA, torch 2.14.1+cu130, 2026-11-08Portablefp16210 3,380 MiB/3,476 MiB–336x518
Kernelsfp16150 3,380 MiB/3,494 MiB–336x518
NVIDIA RTX PRO 6000 CUDA, torch 2.14.1+cu130, 2026-11-08Portablefp16110 3,404 MiB/3,556 MiB–336x518
Kernelsfp1687.6 3,404 MiB/3,534 MiB–336x518

CUDA memory is torch’s peak allocated / reserved; Apple has no such counter, so the Mac rows show the process footprint (weights, Python and torch included). “Output vs reference” compares the device’s output with the stored golden output under the card’s tolerance; – = not checked on that device. Kernels “not implemented” = no kernels for this model on that backend yet: it runs portable there.

Licence and weights

Non-commercial only

Code
LicenseRef-FAIR-Noncommercial-Research LICENSE
Weights
LicenseRef-FAIR-Noncommercial-Research source
Training data
Per the repository README: a mixture of public multi-view datasets (uCO3D annotations released); DINOv3 initialisation. The 512 checkpoint is recommended for in-the-wild use, not benchmarking (vggt_omega_1b_416_reproduce.pt is the benchmark reference).
  • code: LicenseRef-FAIR-Noncommercial-Research (noncommercial)
  • weights: LicenseRef-FAIR-Noncommercial-Research
  • Code and weights are under the FAIR Noncommercial Research License; the checkpoint is gated on Hugging Face (accept the licence there) and not mirrored.

Our reading of the licences, not legal advice.

Weights

vggt_omega_1b_512.pt 4.6 GB· hosted on Hugging Face

fastforks downloads and checks these files on first use (fastforks download vggt-omega-1b-512). Non-commercial weights need FASTFORKS_NONCOMMERCIAL=1.

Accuracy

The test set through the authors’ code and through ours: same checkpoint, same protocol. The distance between the original’s fp32 and fp16 runs is what half precision alone costs; our runs should stay within it.

DTU scans 1

9, 24, 33, 75, view 24 + 7 sources, AbsRel / delta1 %

RundtypeAbsRel ↓δ1.25 % ↑Δ AbsRel vs fp32 originalscansms / scan
Original fp32fp320.005099.93 reference 5 2658
Original fp16bf160.005199.93 −0.0001 5 914
Portable fp32fp320.005099.93 ±0.0000 5 2674
Portablefp160.005099.93 ±0.0000 5 729
Kernelsfp160.005199.93 −0.0001 5 465

640x512 crops, the 7 most co-visible views as sources; GT = depth rendered from the structured-light points. AbsRel / delta1 over GT pixels, after per-image median scaling unless the model takes the cameras (calibrated MVS: cameras, error in mm; PatchMatchNet also the depth range 425-935 mm, MVSAnywhere its own range from the camera baselines). Original = the authors’ code, same checkpoint and protocol. Measured on original fp32, original fp16, portable, kernels: RTX 5060 Ti, torch 2.14.1+cu130, 2026-11-06; portable fp32: RTX 5060 Ti, torch 2.14.1+cu130, fastforks 30c7a8e9, 2026-11-18.

Closeness to the original

Beyond the test-set scores: the raw output of ours against the original’s fp32 output on the same input, pixel by pixel. CPU, float32: max |diff| <= 0.002 vs the original. GPU, float16 (the default), portable and kernels: mean |diff| <= 0.02 and p99 <= 0.25 (absolute error of the output depth).

CheckToleranceMeasured
CPU fp32, max |diff|≤ 0.002 not recorded
GPU default dtypes, mean |diff|≤ 0.02 0.0002
GPU default dtypes, p99 |diff|≤ 0.25 0.00065

Recorded for vggt-omega-1b-512 on RTX 5060 Ti, torch 2.14.1+cu130, 2026-10-31, with these kernels: fastbbvggt_omega

Details

VGGT-Omega (Meta), the successor of VGGT: cameras and depth maps of every view from unposed images in one forward pass. A DINOv3 ViT-L/16 patch embedding feeds 24 frame and 24 inter-frame (global or register-only) attention blocks with q/k norm and 2-D RoPE; a camera head and a DPT dense head decode the tokens. The checkpoint is gated on Hugging Face (accept the FAIR licence there, then put the file on FASTFORKS_WEIGHTS_PATH); there is no mirror.

Input / output
Input: several images of one scene (posed and calibrated for multi-view stereo, unposed otherwise). Output: depth maps and, where the model predicts them, cameras and 3-D points.
Input size
any H x W >= 1, resized or padded inside the model; tested up to 3000x4000 (5,594 MiB allocated)
Options
  • area_side=512>= 16· changes the output
Outputs
depth, confidence, intrinsics, cam_to_world
CUDA kernels
  • fastbbfast backbone kernels: flash attention, window attention, instance norm and rotary embeddings for ViT and similar backbones
  • vggt_omegaVGGT-Omega: fused LayerNorm, residual and rotary kernels
Apple (MPS)
fp32, fp16
Full reference, generated from the repository

VGGT-Omega (Meta): the successor of VGGT with a DINOv3 backbone, cameras and depth from many images. Non-commercial licence.

Commercial use: no (code: LicenseRef-FAIR-Noncommercial-Research (noncommercial); weights: LicenseRef-FAIR-Noncommercial-Research).

variantinputbackendtime (ms)peak memory (MiB, allocated / reserved)
vggt-omega-1b-512synthetic (336x518)original (fp32)
original
portable
kernels
portable + torch.compile
TensorRT fp16
2650
906
726
467
589
377
5765 / 5912
7495 / 7654
3421 / 3518
3404 / 3494
3082 / 3154
5657 / 5766
Details
  • Original code: https://github.com/facebookresearch/vggt-omega at 48b23c8
  • Code licence: LicenseRef-FAIR-Noncommercial-Research (noncommercial); files: LICENSE
  • Weights licence: LicenseRef-FAIR-Noncommercial-Research
  • Licence evidence: https://github.com/facebookresearch/vggt-omega/blob/48b23c8/LICENSE and https://huggingface.co/facebook/VGGT-Omega/blob/1041e80fc0e911235d3426b0a3d9a81075111a53/LICENSE.txt
  • Training data: Per the repository README: a mixture of public multi-view datasets (uCO3D annotations released); DINOv3 initialisation. The 512 checkpoint is recommended for in-the-wild use, not benchmarking (vggt_omega_1b_416_reproduce.pt is the benchmark reference).
  • Licence note: Code and weights are under the FAIR Noncommercial Research License; the checkpoint is gated on Hugging Face (accept the licence there) and not mirrored.
  • Options: area_side = 512 (>= 16; also resolution)
  • Kernel libraries: fastbb, vggt_omega
  • Variants: vggt-omega-1b-512
  • Outputs: depth, confidence, intrinsics, cam_to_world; kind: scale_invariant
  • Input size: any H x W >= 1, resized or padded inside the model
  • Tested up to (portable, fp16, RTX 5060 Ti 16 GB): vggt-omega-1b-512 3000x4000 (5.5 GiB peak allocated, 6.7 reserved)
  • vggt-omega-1b-512 336x518, first call: portable + torch.compile 66 s (cold Inductor cache), TensorRT fp16 187 s (ONNX + engine).
  • vggt-omega-1b-512 original (fp32) 336x518: float32 (the authors' autocast off)
  • vggt-omega-1b-512 original 336x518: original default VGGTOmega(autocast=True): bfloat16 aggregator, float32 heads
  • vggt-omega-1b-512 load / first call: original 16.31 s / 1.1 s (336x518), portable 3.31 s / 0.734 s (336x518), kernels 5.24 s / 0.752 s (336x518)
  • Protocol, median of 9 after 3 warm-up, uint8 CUDA input (synthetic for bench_inputs sizes), card defaults; gc before each load; peak = max over the first call and the timed calls

Measured

640x512 crops, the 7 most co-visible views as sources; GT = depth rendered from the structured-light points. AbsRel / delta1 over GT pixels, after per-image median scaling unless the model takes the cameras (calibrated MVS: cameras, error in mm; PatchMatchNet also the depth range 425-935 mm, MVSAnywhere its own range from the camera baselines). Original = the original code named under Details, same checkpoint and protocol.

variantdatasetoriginal fp32original fp16portableportable fp32kernelsnote
vggt-omega-1b-512DTU scans 1, 9, 24, 33, 75, view 24 + 7 sources, AbsRel / delta1 %0.0050 / 99.930.0051 / 99.930.0050 / 99.930.0050 / 99.930.0051 / 99.93

Measured on portable: NVIDIA GeForce RTX 5060 Ti, torch 2.14.1+cu130, 2026-11-06; portable32: NVIDIA GeForce RTX 5060 Ti, torch 2.14.1+cu130, fastforks 30c7a8e9, 2026-11-18; kernels: NVIDIA GeForce RTX 5060 Ti, torch 2.14.1+cu130, 2026-11-06.

Parity

CPU, float32: max |diff| <= 0.002 vs the original. GPU, float16 (the default), portable and kernels: mean |diff| <= 0.02 and p99 <= 0.25 (absolute error of the output depth).

variantkernels error vs original, mean / p99GPU, torchkernels used
vggt-omega-1b-5120.0002 / 0.00065RTX 5060 Ti, torch 2.14.1+cu130fastbb@7d058322634d:sm_120, vggt_omega@b8fa01afd00c:sm_120

Example output

One fixed input through this checkpoint, the same for every model of the task (kernels · float16 · NVIDIA GeForce RTX 5060 Ti).

View 1
View 2
Output
Motorcycle, Middlebury 2014 stereo (left and right view, downscaled 4x), view 1
View 1 · 100%

Input: Motorcycle, Middlebury 2014 stereo (left and right view, downscaled 4x), Scharstein et al., Middlebury College (research dataset, no licence stated).