Models · Multi-view depth

AnySplat

AnySplat (InternRobotics): 3D Gaussian splats, cameras and depth from unposed images in one pass; saves a 3DGS .ply. MIT.

experimental Fork of InternRobotics/AnySplat@5f5e208 · backends: portable, kernels

Checkpoint anysplat

Paper 2025 Added in 0.2.0 Evaluated with 0.2.0

Kernels, RTX 5060 Ti 16 GB 364 ms 336x518, fp16
vs the faster original 2.0× original bf16: 741 ms
Peak memory 4,849 MiB original: 8,931 MiB
Accuracy vs original fp32 ±0.0000 kernels AbsRel, DTU scans 1 (0.0108 original)

Use it

pip install fastforks

Lean install with only what this model needs: pip install fastforks-core

from fastforks import multiview

model = multiview.load("anysplat")  # kernels on a supported GPU, else portable PyTorch
result = model(images)
result.depth  # also: confidence, world_points, intrinsics, cam_to_world, gaussians
result.backend  # what ran: mode, device, dtype, kernels
fastforks bench anysplat

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 fp32fp32not measured ––––
Original (own default)the authors' mixed-precision option (bfloat16); InternRobotics/AnySplat@5f5e208a7dd5 EncoderAnySplat inference, long side 448, voxel_size 0.002bf16 741–8,931 MiB/9,394 MiB7.00.82
Portable fp32fp32not measured ––––
Portablefp16 5031.5×4,842 MiB/4,932 MiB4.70.89
Kernelsfp16 3642.0×4,849 MiB/5,192 MiB4.00.71
Portable + torch.compile 4271.7×4,137 MiB/4,694 MiB80 (build)–
TensorRT fp16fp16 3142.4×4,504 MiBenginenot comparable215 (build)–

Measured on RTX 5060 Ti 16 GB, torch 2.14.1+cu130, 2026-11-12 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, 2026-11-14Originalfp32cannot run here–336x518
Portablefp165444 8,751 MiB processmatches336x518
Portablefp3233262 10.5 GiB processmatches336x518
Kernelsfp16not implemented–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

Commercial use allowed

Code
MIT AND LicenseRef-VGGT LICENSELICENSE-VGGT
Weights
MIT source
Training data
Per the paper (arXiv 2505.23716) a mixture of public multi-view datasets; the repository ships CO3Dv2, DL3DV and ScanNet++ training configs. VGGT-1B initialisation.
  • AnySplat's code and checkpoint are MIT; its VGGT copy (net/) is under the VGGT License (commercial use allowed). The checkpoint was fine-tuned from VGGT-1B, whose own weights are CC-BY-NC-4.0: the AnySplat authors release theirs as MIT.
  • Upstream's Gaussian head file (vggt_dpt_gs_head.py) carries a stale Naver CC BY-NC-SA header from DUSt3R although the class used here is new code on the VGGT DPT head; this family reimplements that head (GSHead in model.py) and vendors none of the Naver-headed files.

Our reading of the licences, not legal advice.

Weights

model.safetensors 2.9 GB· hosted on Hugging Face

fastforks downloads and checks these files on first use (fastforks download anysplat).

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 fp32fp320.010899.93 reference 5 2352
Original fp16bf160.010899.93 ±0.0000 5 956
Portable fp32fp320.010899.93 ±0.0000 5 2345
Portablefp160.010999.93 −0.0001 5 645
Kernelsfp160.010899.93 ±0.0000 5 461

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: RTX 5060 Ti, torch 2.14.1+cu130, 2026-11-12; portable fp32, portable, kernels: RTX 5060 Ti, torch 2.14.1+cu130, fastforks 62af007c, 2026-11-12.

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| <= 1.3e-05 vs the original. GPU, float16 (the default), portable and kernels: mean |diff| <= 0.0013 and p99 <= 0.0033 (absolute error of the output depth).

CheckToleranceMeasured
CPU fp32, max |diff|≤ 0.000013 not recorded
GPU default dtypes, mean |diff|≤ 0.0013 0.00082
GPU default dtypes, p99 |diff|≤ 0.0033 0.00214

Recorded for anysplat on RTX 5060 Ti, torch 2.14.1+cu130, 2026-11-12, with these kernels: cudnn_haccfastbb

Details

AnySplat (Jiang et al. 2025): feed-forward 3D Gaussian splatting from unposed, uncalibrated images. A VGGT backbone (fine-tuned aggregator, camera head and depth head) predicts the cameras and depth of every view; a DPT Gaussian head predicts per-pixel Gaussians (opacity, scale, rotation, degree-4 SH), fused per voxel into one scene (`result.gaussians`, `.save("scene.ply")`). Forward pass only: no renderer.

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 (6,627 MiB allocated)
Options
  • long_side=448>= 14· changes the output
  • voxel_size=0.002>= 0.0· changes the output
Outputs
depth, confidence, world_points, intrinsics, cam_to_world, gaussians
CUDA kernels
  • fastbbfast backbone kernels: flash attention, window attention, instance norm and rotary embeddings for ViT and similar backbones
  • cudnn_haccconvolutions with fp16 accumulation (cuDNN)
Apple (MPS)
fp32, fp16

Limitations

  • Kernels peak allocated is equal to portable, but the cuDNN workspace next to the large conv output (257 MiB in the DPT head) fragments the allocator: peak reserved is ~260 MiB above portable.
Full reference, generated from the repository

AnySplat (InternRobotics): 3D Gaussian splats, cameras and depth from unposed images in one pass; saves a 3DGS .ply. MIT.

Commercial use: yes.

variantinputbackendtime (ms)peak memory (MiB, allocated / reserved)
anysplatsynthetic (336x518)original
portable
kernels
portable + torch.compile
TensorRT fp16
741
503
364
427
314
8931 / 9394
4842 / 4932
4849 / 5192
4137 / 4694
7629 / 7674
Details
  • Original code: https://github.com/InternRobotics/AnySplat at 5f5e208
  • Code licence: MIT AND LicenseRef-VGGT (permissive); files: LICENSE, LICENSE-VGGT
  • Weights licence: MIT
  • Licence evidence: https://github.com/InternRobotics/AnySplat/blob/5f5e208a7dd57d52e43ea0d553a95eab526e8775/LICENSE and https://huggingface.co/lhjiang/anysplat/blob/d2e8c343672646041ad4ea518184968f94362f01/README.md (license: mit)
  • Training data: Per the paper (arXiv 2505.23716) a mixture of public multi-view datasets; the repository ships CO3Dv2, DL3DV and ScanNet++ training configs. VGGT-1B initialisation.
  • Licence note: AnySplat's code and checkpoint are MIT; its VGGT copy (net/) is under the VGGT License (commercial use allowed). The checkpoint was fine-tuned from VGGT-1B, whose own weights are CC-BY-NC-4.0: the AnySplat authors release theirs as MIT.
  • Licence note: Upstream's Gaussian head file (vggt_dpt_gs_head.py) carries a stale Naver CC BY-NC-SA header from DUSt3R although the class used here is new code on the VGGT DPT head; this family reimplements that head (GSHead in model.py) and vendors none of the Naver-headed files.
  • Options: long_side = 448 (>= 14), voxel_size = 0.002 (>= 0.0)
  • Kernel libraries: fastbb, cudnn_hacc
  • Variants: anysplat
  • Outputs: depth, confidence, world_points, intrinsics, cam_to_world, gaussians; 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): anysplat 3000x4000 (6.5 GiB peak allocated, 9.4 reserved)
  • anysplat 336x518, first call: portable + torch.compile 80 s (cold Inductor cache), TensorRT fp16 215 s (ONNX + engine).
  • anysplat original 336x518: the authors' mixed-precision option (bfloat16); InternRobotics/AnySplat@5f5e208a7dd5 EncoderAnySplat inference, long side 448, voxel_size 0.002
  • anysplat load / first call: original 8.32 s / 3.068 s (336x518), portable 4.71 s / 0.891 s (336x518), kernels 4.04 s / 0.708 s (336x518)
  • Protocol, first call, 3 warm-up, peak reset, median of 3 rounds x median of 5

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
anysplatDTU scans 1, 9, 24, 33, 75, view 24 + 7 sources, AbsRel / delta1 %0.0108 / 99.930.0108 / 99.930.0109 / 99.930.0108 / 99.930.0108 / 99.93

Measured on NVIDIA GeForce RTX 5060 Ti, torch 2.14.1+cu130, fastforks 62af007c, 2026-11-12.

Parity

CPU, float32: max |diff| <= 1.3e-05 vs the original. GPU, float16 (the default), portable and kernels: mean |diff| <= 0.0013 and p99 <= 0.0033 (absolute error of the output depth).

variantkernels error vs original, mean / p99GPU, torchkernels used
anysplat0.00082 / 0.00214RTX 5060 Ti, torch 2.14.1+cu130cudnn_hacc@cudnn92400, fastbb@7d058322634d:sm_120

Limitations

  • Kernels peak allocated is equal to portable, but the cuDNN workspace next to the large conv output (257 MiB in the DPT head) fragments the allocator: peak reserved is ~260 MiB above portable.

Example output

No example rendered for this checkpoint yet.