MASHBOXLabs
Request access

Lamproite Packs

Each model below is available as a Pack. You run it in your own Google Cloud project, or we host it for you as a private endpoint.

What a Pack gives you

These models are shown as premade examples, not on the public API. To evaluate one, write to [email protected].

Z-Image Turbo

Model Available as a Pack

Text to image.

text in, image out. By Tongyi-MAI (Alibaba).

“A translucent jellyfish-like alien floating above a frozen lake under the aurora”
“A translucent jellyfish-like alien floating above a frozen lake under the aurora”
“A dragon made of volcanic rock with glowing lava cracks, perched on a cliff at dusk”
“A dragon made of volcanic rock with glowing lava cracks, perched on a cliff at dusk”
“Close-up of a hummingbird drinking from a red flower, shallow depth of field”
“Close-up of a hummingbird drinking from a red flower, shallow depth of field”

More about Z-Image Turbo

Wan 2.2 Image to Video

Model Available as a Pack

An image and a prompt in, a video clip out.

image + text in, video out. By Wan team (Alibaba).

From a photo and the prompt “a group of camels resting in the desert”. 81 frames at 832x464.
From a photo and the prompt “a table and chairs in a room with sunlight coming through the window”. 81 frames at 832x464.

More about Wan 2.2 Image to Video

Qwen3-30B-A3B

Model Available as a Pack

Chat with streaming, OpenAI-compatible.

text in, text out. By Qwen team (Alibaba).

More about Qwen3-30B-A3B

Track2Rig

Model Experimental Available as a Pack

3D point tracks of a moving body in; skeleton, skin and bone motion out as GLB.

3D point tracks in, skeleton, skin and motion (GLB) out. By MASHBOX Labs.

3D point tracks of a walking creature it never saw in; skeleton (black) and skin (colours) out.

More about Track2Rig

MASHBOX Stride

Model Available as a Pack

Describe a movement and get a humanoid motion clip.

text in, motion clip out. By MASHBOX Labs.

“someone waves both hands above their head excitedly”
“a person does a threatening crouching walk and then raises their arms”
“a person jumps up into the air, turns his body three quarters of the way around and lands on his feet”

More about MASHBOX Stride

TokenRig

Model Available as a Pack

Auto-rig and skin any mesh.

3D mesh in, rigged mesh out. By VAST.

A humanoid mesh rigged automatically: skeleton front and side, and skin weights by bone.
A humanoid mesh rigged automatically: skeleton front and side, and skin weights by bone.

More about TokenRig

Audio2Face-3D

Model Available as a Pack

Speech audio to facial animation.

speech audio in, facial animation out. By NVIDIA.

More about Audio2Face-3D

Granite Speech 4.1 2B

Model Available as a Pack

Speech to text, OpenAI-compatible.

speech audio in, text out. By IBM.

Input: LibriSpeech test-clean, CC BY 4.0.
Our transcriptShe pushed him toward the big chair by the fire and sat down on a stool at the opposite side of the hearth, her knees drawn up to her chin, laughing like a happy little girl.

More about Granite Speech 4.1 2B

Z-Image ControlNet Union

Model Available as a Pack

Images guided by a pose, depth or edge image.

text + control image in, image out. By Alibaba PAI.

Black-and-white photo (sample control image)
Black-and-white photo (sample control image)
“A color photograph of a man in a white t-shirt and khaki shorts walking along a curving wooden boardwalk through lush green tropical trees, apartment towers in the hazy distance, overcast sky.”
“A color photograph of a man in a white t-shirt and khaki shorts walking along a curving wooden boardwalk through lush green tropical trees, apartment towers in the hazy distance, overcast sky.”
Edge map (sample control image)
Edge map (sample control image)

More about Z-Image ControlNet Union

Upscaler

Model Available as a Pack

Image and video upscaling.

image or video in, larger image or video out. By MASHBOX Labs.

510x339 photo with noise and blur
510x339 photo with noise and blur
4x: 2040x1356
4x: 2040x1356

More about Upscaler

AniGen

Model Available as a Pack

An image to a rigged 3D model.

image in, rigged 3D model out. By VAST.

More about AniGen