CPU3DAI tools for 3D, video and audio

AI tools under 8 GB of VRAM

VRAM thresholds, backends and licenses come from the tool fact sheets: every value is collected by code from an official source and stamped with the date it was checked. Requirements marked with an asterisk * are the authors’ own description, not our measurement. Details are on each tool’s page.

Showing 16 of 116

ToolTypeRunsVRAMBackendsFormats
LATO.2image-to-3d3Dopen sourcefrom 8 GB *CUDAobj, ply
MeshAnything3Dopen sourcefrom 7 GB *CUDA—
Stable Fast 3Dimage-to-3d3Dopen sourcefrom 6 GB *CUDAApple MPSglb
threestudio3Dopen sourcefrom 6 GB *CUDAobj
TripoSRimage-to-3d3Dopen sourcefrom 6 GB *CUDA—
Zero123++3Dopen sourcefrom 5 GB *——
ACE-Steptext-to-audioAudioopen sourcefrom 8 GB *——
ChatTTStext-to-audioAudioopen sourcefrom 4 GB *——
DiffRhythmAudioopen sourcefrom 8 GB *——
MMAudioAudioopen sourcefrom 6 GB *CUDA—
MOSS-SoundEffect without CUDA (Mac and AMD)Audioopen sourcefrom 8 GB (8–10 GB) *Apple MPSAMD ROCm—
FramePackVideoopen sourcefrom 6 GB *CUDA—
LTX-Videoimage-to-videoVideoopen sourcefrom 1 GB *Apple MPS—
Maestroimage-text-to-videoVideoopen sourcefrom 6 GB (6–24 GB) *CUDA—
Pyramid Flowtext-to-videoVideoopen sourcefrom 8 GB *Apple MPS—
SuperMimic-ReproducedVideoopen sourcefrom 6.7 GB *——

Cloud services — no VRAM needed

They compute on their own servers, so a hardware filter does not apply to them.

KaedimTurns sketches, references, product photos and briefs into 3D assets3Dcloudcloudnot needed—
Luma GenieGeneration and processing of video, images, audio and text3Dcloudcloudnot neededJPG
Masterpiece XAI editor for creating 3D scenes and assets3Dcloudcloudnot needed—
Meshy3D model generation from text and images, AI texturing, animation, rigging3Dcloudcloudnot neededGLB, FBX, OBJ, USDZ, STL
Polycam3D scanning, floor plan creation, reality capture with AI3Dcloudcloudnot neededGLTF, OBJ, FBX, DAE, STL, USDZ, …
Rodin3D model generation from text and images3Dcloudcloudnot neededSTL, FBX, OBJ, GLB, GLTF, USDZ
Tripo AItext-to-3D and image-to-3D3Dcloudcloudnot neededGLB, FBX, OBJ, USD, STL, 3MF
Dream MachineMultimodal generation and editing of images, video, audio and textVideocloudcloudnot neededMP4
HailuoVideo and image generationVideocloudcloudnot needed—
KlingVideo and image generationVideocloudcloudnot needed—
PikaAI generation of video, images and audioVideocloudcloudnot needed—
RunwayGeneration of video, images, speech, sound effects, avatars, plus video and image processingVideocloudcloudnot neededMP4, MOV
SoraVideo generation from text, images or videoVideocloudcloudnot neededMP4
VeoVideo generationVideocloudcloudnot needed—
ViduVideo generation from text, images and references; interactive digital humansVideocloudcloudnot needed—

Requirements not stated by the authors

The official sources for these tools have no data for the selected filter — that means “unknown”, not “won’t work”.

3DTopia3Dopen sourcenot statedCUDAglb
CRM3Dopen sourcenot stated—obj
DreamGaussian3Dopen sourcenot statedCUDA—
Era3D3Dopen sourcenot stated——
FieldCraft3Dopen sourcenot statedApple MPS3mf
Gaussian-Splat-Lite3Dopen sourcenot stated—ply
Hi3D v1.0image-to-3d3Dopen sourcenot statedCUDA—
Hunyuan3D 2.1 without CUDA (Mac and AMD)3Dopen sourcenot statedApple MPSAMD ROCmobj, glb, ply, stl, fbx, dae, 3mf
idolimage-to-3d3Dopen sourcenot stated——
InstantMeshimage-to-3d3Dopen sourcenot statedCUDAobj
LichtFeld-Studio3Dopen sourcenot statedCUDAply
Michelangelo3Dopen sourcenot stated——
MVDream3Dopen sourcenot stated——
Nerfstudio3Dopen sourcenot statedCUDA—
One-2-3-453Dopen sourcenot stated——
OOOSplat3Dopen sourcenot statedApple MPSply
OpenLRMimage-to-3d3Dopen sourcenot statedCUDAobj
PhysX-Omniimage-to-3d3Dopen sourcenot stated——
Point-E3Dopen sourcenot stated——
Scal3Rimage-to-3d3Dopen sourcenot statedCUDA—
Shap-E3Dopen sourcenot stated——
TRELLISimage-to-3d3Dopen sourcenot statedCUDAglb
Unique3D3Dopen sourcenot statedCUDA—
video_to_world3Dopen sourcenot statedCUDAply
Wonder3D3Dopen sourcenot stated——
CosyVoicetext-to-speechAudioopen sourcenot stated——
F5-TTStext-to-speechAudioopen sourcenot statedAMD ROCm—
Fish Speechtext-to-speechAudioopen sourcenot statedCUDA—
MOSS-SoundEffecttext-to-audioAudioopen sourcenot stated——
MusicGentext-to-audioAudioopen sourcenot stated——
OpenVoicetext-to-speechAudioopen sourcenot stated——
Stable Audio Opentext-to-audioAudioopen sourcenot stated——
XTTStext-to-speechAudioopen sourcenot statedCUDA—
AI Comic BuilderVideoopen sourcenot stated——
AlayaRendererimage-to-imageVideoopen sourcenot stated——
BlinkMimicVideoopen sourcenot statedCUDA—
CogVideoXtext-to-videoVideoopen sourcenot statedCUDA—
Cosmos3-Super-Image2Videoimage-to-videoVideoopen sourcenot statedCUDA—
DataMagicVideoopen sourcenot stated——
doubao2apiVideoopen sourcenot stated——
EdenLiveimage-to-videoVideoopen sourcenot statedCUDA—
EMERGE-Policyimage-to-3dVideoopen sourcenot stated——
forge-filmVideoopen sourcenot statedCUDA—
frametellVideoopen sourcenot statedApple MPS—
GAE-GeometricAutoEncoderimage-to-videoVideoopen sourcenot statedCUDAply
Infinite WorldVideoopen sourcenot stated——
LatteVideoopen sourcenot statedCUDA—
LingBot-VideoVideoopen sourcenot stated——
MACaQVideoopen sourcenot stated——
nano-world-modelVideoopen sourcenot statedCUDA—
object-permanencevideo-to-videoVideoopen sourcenot stated——
Open-SoraVideoopen sourcenot stated——
OpenSplineVideoopen sourcenot statedCUDA—
RefAlignimage-text-to-videoVideoopen sourcenot stated——
ShrimpReelVideoopen sourcenot stated——
snapcnVideoopen sourcenot stated——
SparkDiffusiontext-to-videoVideoopen sourcenot statedCUDA—
Stable Video Diffusionimage-to-videoVideoopen sourcenot stated——
TE-Speed-MiniMaxH3-OSSVideoopen sourcenot stated——
video-generator-clientVideoopen sourcenot stated——
VideoCrafterVideoopen sourcenot stated——
ViMaxVideoopen sourcenot stated——
ViralForgeVideoopen sourcenot stated——
WHITEBOARD ANIMATION VIimage-segmentationVideoopen sourcenot statedCUDA—
wind-comicVideoopen sourcenot stated——
WorldCrafterimage-to-videoVideoopen sourcenot statedCUDA—

Fact sheets are re-checked monthly; every value on a tool page has a source link and a check date. The asterisk * means “as described by the author”: a developer’s statement, not an independent measurement.