Models

Interpolation models

10 interpolation models. Each section says what the model is, whether it may be used commercially and how fast it is; licences and checks are under Details.

Input: two RGB images (consecutive frames). Output: the frame between them at timestep (0.5 by default; 0 = the first, 1 = the second), RGB in [0, 1].

Measured on RTX 5060 Ti, torch 2.14.1+cu130; fastforks bench MODEL --size HxW measures yours. The summary uses one input, 1080x1920: kernels time with its speed-up over the faster original run, and kernels peak reserved memory with its ratio to the smaller original one. Where measured (one variant per family), a variant's table also lists portable under torch.compile and the fp16 TensorRT engine of portable (export; memory: the engine's device memory + torch's peak). – = not measured; no kernels = the family has no kernels backend (reason under its table). Reserved memory is the allocator's cache: kernels (CUDA graphs, cuDNN workspaces) can reserve a few percent to some tens of percent more than portable at the same allocated peak.

Other devices: Apple M1 Pro 16 GB, NVIDIA L4, NVIDIA L40S, NVIDIA A10, NVIDIA A100-SXM4-40GB, NVIDIA RTX PRO 6000; across devices.

modelcommercialkernels time (ms),
speed-up vs original
kernels peak memory (MiB, reserved),
ratio to original
ema-vfiyes207, 1.4x3566, 0.83x
ema-vfi-smallyes84.4, 1.4x1914, 0.8x
ema-vfi-small-tyes84.3, 1.4x1914, 0.8x
ema-vfi-tyes207, 1.4x3566, 0.83x
filmyes332, 1.4x3080, 0.87x
gimm-vfi-rno16295362
gimm-vfi-r-pno16285282
rife-4.25yes21.4, 1.1x916, 0.79x
rife-4.25-liteyes18.3, 1.2x896, 0.74x
rife-4.26yes21.4, 1.1x916, 0.79x

All models

Modes, licences, training data and outputs per model

Commercial status per variant: yes if the code and the weights licence both allow commercial use (the training data is listed but does not change it; weights without a licence of their own are assumed to carry the code licence); no if either forbids it; unclear for custom licence terms (treat as no).

modelmodescode licenceweights licencetraining datacommercialoutputs
ema-vfiportable, kernelsApache-2.0Apache-2.0vimeo90kyesframe
ema-vfi-smallportable, kernelsApache-2.0Apache-2.0vimeo90kyesframe
ema-vfi-small-tportable, kernelsApache-2.0Apache-2.0vimeo90kyesframe
ema-vfi-tportable, kernelsApache-2.0Apache-2.0vimeo90kyesframe
filmportable, kernelsApache-2.0Apache-2.0vimeo90kyesframe
gimm-vfi-rportable, kernelsLicenseRef-noncommercialLicenseRef-noncommercialvimeo90kno (code: LicenseRef-noncommercial (noncommercial); weights: LicenseRef-noncommercial)frame
gimm-vfi-r-pportable, kernelsLicenseRef-noncommercialLicenseRef-noncommercialvimeo90kno (code: LicenseRef-noncommercial (noncommercial); weights: LicenseRef-noncommercial)frame
rife-4.25portable, kernelsMITMITvimeo90kyesframe
rife-4.25-liteportable, kernelsMITMITvimeo90kyesframe
rife-4.26portable, kernelsMITMITvimeo90kyesframe

Training datasets

iddatasetlicencecommercial use
vimeo90kVimeo-90K (septuplets / triplets for frame interpolation)not statedunclear