Does Vast.ai charge egress or data transfer fees?
💡 Answer
Egress fees at Vast.ai: Varies by host ($/TB)
Egress fees are the charges applied when you transfer data out of the cloud provider (e.g., downloading trained model weights, serving inference results, or moving datasets to another provider). This is an important cost consideration for ML workflows that involve frequent model exports or large dataset movements.
Storage options: Varies by host ($/GB/hr, charged while instance exists)
For the complete data transfer fee schedule and free egress tiers, see Vast.ai official website.
More FAQs about Vast.ai
- What is Vast.ai best for?
- What is the current Trustpilot rating and number of reviews for Vast.ai?
- What machine learning frameworks does Vast.ai support?
- How fast can I deploy a GPU instance on Vast.ai?
- Does Vast.ai offer serverless GPU inference?
- Where are Vast.ai data centers located?
- Does Vast.ai support multi-GPU instances with NVLink or InfiniBand?
- Does Vast.ai offer spot or preemptible GPU instances?
- Does Vast.ai offer free credits or a free trial?
- What GPU models does Vast.ai offer?
- What is Vast.ai pricing and how does billing work?
Guides Where Vast.ai Is Featured
- Best Cloud GPU Providers with AMD MI300X
- Best Cloud GPUs for Research & Experimentation
- Cheapest Cloud GPUs Under $1/hr
- Cloud GPU Providers with API & CLI Management
- Cloud GPU Providers with Docker & Custom Images
- Cloud GPU Providers with Free Credits
- Cloud GPU Providers with Jupyter Notebook Support
- Cloud GPU Providers with Kubernetes Support
- Cloud GPU Providers with Multi-Node GPU Clusters
- Cloud GPU Providers with NVLink or InfiniBand
- Cloud GPU Providers with Per-Second Billing
- Cloud GPU Providers with Persistent Storage
- Cloud GPU Providers with Serverless GPU Inference
- Cloud GPU Providers with Spot / Preemptible Instances
- Cloud GPU Providers with SSH Access
- Cloud GPU Providers with Zero Egress Fees
These guides include Vast.ai alongside other cloud GPU providers, grouped by hardware, pricing, features, and infrastructure.
Vast.ai vs Latitude.sh vs Massed Compute - GPU Provider Comparison (March 2026)
Side-by-side comparison of Vast.ai vs Latitude.sh vs Massed Compute. Quickly scan maximum funding, profit splits, risk rules, leverage, platforms, instruments, payout schedules, payment options, trading permissions and KYC restrictions to narrow down your prop trading firm shortlist. Data updated March 2026.
|
Vast.ai
Instant GPUs. Transparent Pricing.
|
Latitude.sh
Bare metal GPU cloud across 23 global locations
|
Massed Compute
GPU cloud with direct engineer support
|
|
|---|---|---|---|
| Overview | |||
| Trustpilot Rating | 4.4 | 3.7 | 0 |
| Headquarters | United States | Brazil | United States |
| Provider Type | GPU Marketplace | Bare Metal | GPU-Focused |
| Best For | AI training inference fine-tuning Stable Diffusion batch processing research LLM serving generative AI | AI training inference bare metal GPU fine-tuning research dedicated workloads generative AI | AI training inference VFX rendering generative AI fine-tuning HPC Stable Diffusion research |
| GPU Hardware | |||
| GPU Models | B200 H200 H100 SXM H100 NVL A100 SXM A100 PCIe RTX 5090 RTX 5080 RTX 5070 Ti RTX 6000 Pro RTX 6000 Ada RTX 4500 Ada RTX A6000 RTX A5000 RTX A4000 L40S L40 A40 A10 RTX 4090 RTX 4080 RTX 4070 Ti RTX 4070 RTX 4060 Ti RTX 4060 RTX 3090 Ti RTX 3090 RTX 3080 Ti RTX 3080 RTX 3070 Ti RTX 3070 Tesla V100 Tesla T4 A2 GTX 1080 | A30 RTX A5000 RTX A6000 L40S RTX 6000 Ada A100 SXM H100 SXM GH200 RTX PRO 6000 | A30 RTX A5000 RTX A6000 L40S A100 SXM H100 PCIe H100 SXM H100 NVL RTX PRO 6000 H200 NVL |
| Max VRAM (GB) | 192 | 96 | 141 |
| Max GPUs/Instance | 8 | 8 | 8 |
| Interconnect | NVLink, InfiniBand | NVLink | NVLink |
| Pricing | |||
| Starting Price ($/hr) | $0.06/hr | $0.35/hr | $0.35/hr |
| Billing Granularity | Per-second | Per-hour | Per-minute |
| Spot/Preemptible | 1 | 0 | 0 |
| Reserved Discounts | Up to 50% (1-6 month reserved) | N/A | N/A |
| Free Credits | Small test credit on signup | $200 via referral program | None |
| Egress Fees | Varies by host ($/TB) | None | None |
| Storage | Varies by host ($/GB/hr, charged while instance exists) | Local NVMe included (up to 4x 3.8TB), Block Storage $0.10/GB/mo, Filesystem Storage $0.05/GB/mo | Local NVMe included with instances |
| Infrastructure | |||
| Regions | 500+ locations, 40+ data centers | 23 locations: US (8 cities), LATAM (5), Europe (5), APAC (4), Mexico City. GPU in Dallas, Frankfurt, Sydney, Tokyo | United States (Tier III data centers) |
| Uptime SLA | No formal SLA (host reliability scores visible) | 99.9% | Tier III (99.98% design) |
| Developer Experience | |||
| Frameworks | PyTorch TensorFlow CUDA vLLM ComfyUI | ML-optimized images PyTorch TensorFlow (user-installed) CUDA | PyTorch TensorFlow CUDA cuDNN ComfyUI pre-configured ML templates |
| Docker Support | 1 | 1 | 1 |
| SSH Access | 1 | 1 | 1 |
| Jupyter Notebooks | 1 | 0 | 0 |
| API / CLI | 1 | 1 | 1 |
| Setup Time | Seconds | Seconds | Minutes |
| Kubernetes Support | 0 | 0 | 0 |
| Business Terms | |||
| Min Commitment | None | None | None |
| Compliance | SOC 2 Type 2 HIPAA GDPR CCPA | Single-tenant isolation DPA available | SOC 2 Type II HIPAA |