Deep Learning VM Image
Reviews
Star rating
Rank in Infrastructure as a Service (IaaS)
Pros and cons
Pros
- Ease of Use
- Setup Ease
- AI Integration
- Features
- Integrations
- Easy Setup
Cons
- Expensive
- Cost
- Learning Difficulty
- Difficult Learning
- Limited Customization
- Pricing Issues
Traffic & LLM citations
Source: G2 review data
Winning from
Losing to
Key stats
Adoption by company size
| Segment | Share of reviews |
|---|---|
| Small Business | 47.9% |
| Mid-Market | 35.2% |
| Enterprise | 16.9% |
Quarterly history
| Quarter | Reviews | Star rating |
|---|---|---|
| 2024 Q4 | 44 | 4.4 |
| 2025 Q1 | 2 | 4.8 |
| 2025 Q3 | 3 | 4.5 |
| 2025 Q4 | 1 | 4.0 |
| 2026 Q2 | 1 | 3.5 |
| 2026 Q3 | 19 | 4.2 |
Satisfaction ratings
What reviewers say
Massive time-saver for setting up ML environments, though you need to watch your GPU costs!
“Setting up a deep learning environment from scratch is usually a miserable cycle of matching Linux kernels, NVIDIA drivers, CUDA versions, and PyTorch builds. Google's DL VMs completely wipe out that headache. You spin up the instance, open JupyterLab, and everything actually wor”
Effortless Deep Learning Setup, Minor Learning Curve
“I like Deep Learning VM Image because it comes with pre-installed tools and libraries, making it easy to start AI and machine learning projects without complicated setup. The pre-installed Python libraries like TensorFlow and PyTorch are particularly helpful for building and trai”
GPU JupyterLab Ready Out of the Box, No Driver or CUDA Setup Needed
“It completely skips the nightmare of installing NVIDIA drivers and CUDA dependencies manually. You can launch a GPU-enabled compute Engine instance via gcloud and immediately jump into a running JupyterLab setup with PyTorch or TensorFlow pre- installed. It just works out of the ”