Every package below normally requires hours of compilation from source. We build them against the exact Colab runtime so you don't have to — and we make sure they install cleanly alongside every default Colab package. No uninstalls, no downgrades, no version conflicts. They just work.
Every wheel is monkey-patched and pinned to coexist with the default Colab stack. That means no collisions with preinstalled packages, no forced uninstalls of torch or numpy, and no version battling. You run one pip install and everything just works — your existing imports stay intact.
Optimized Flash Attention 2 — the backbone of efficient transformer inference on A100s.
NVIDIA's differentiable rasterizer for 3D deep learning, prebuilt with CUDA support.
CUDA-accelerated mesh processing for 3D pipeline work. Multi-arch builds for T4, A100 & L4.
CUDA voxel utilities for 3D pipeline work. No build step required.
Flexible GEMM kernels optimized for mixed-precision workloads on Ampere GPUs.
Neural rendering components for reconstruction pipelines, ready to import.
3D math and geometry utilities with CUDA acceleration. No build step required.
Memory-efficient attention and transformer building blocks from Meta. Critical for running large models on limited VRAM.
Run GGUF-quantized Stable Diffusion models on Colab. Compiled for frictionless T4 compatibility — no dependency conflicts with the default stack.
| Spec | Details |
|---|---|
| GPU | NVIDIA A100 NVIDIA L4 NVIDIA T4 |
| Platform | Google Colab linux x86_64 |
| Python | 3.12 |
| CUDA | 12.8 |
| PyTorch | 2.10 |
Pre-configured Colab notebooks that use MissingLink wheels out of the box. Paste your token, hit run, and start producing output — no setup, no debugging, no dependency chasing.
Generate high-quality 3D models (.glb) from images using Microsoft's TRELLIS.2 4B model. Batch processing with automatic MP4 previews and Google Drive resume support.
Fast AI image generation powered by Z-Image. Optimized for A100 with flash attention acceleration.
Z-Image with GGUF quantization for efficient image generation. Lower VRAM usage while maintaining quality.
Text-to-video generation using Wan2.2 with GGUF quantization. Generate video clips from text prompts on Colab.
More notebooks on the way — Support