Novita AI

GPU-Focused Headquartered in United States Founded in 2022
Updated March 19, 2026

Novita AI is a San Francisco-based AI cloud platform founded in 2022 that provides GPU instances, serverless model APIs, and an Agent Sandbox for developers. The platform offers access to 200+ AI models through a single API, covering LLMs, image generation, video, text-to-speech, and embeddings, alongside dedicated GPU cloud instances for training and fine-tuning.

Their GPU cloud features 7 GPU models from RTX 3090 to H100 SXM with up to 8 GPUs per instance, per-second billing, and spot pricing at 50% off on-demand rates. Novita AI operates across 20+ data center locations on 4+ continents with sub-20ms network latency. The platform includes pre-configured environments with PyTorch, TensorFlow, and Jupyter notebooks, plus Docker support for custom images.

Novita AI offers a generous startup program with up to $10,000 in free credits and zero egress fees across all plans. The platform supports both serverless inference with auto-scaling and dedicated GPU instances, making it suitable for everything from prototype development to production-scale AI deployment.

Starting Price $0.11/hr Per hour
Max VRAM 80 GB Per GPU
Max GPUs 8 Per instance
Billing Per-second Granularity

GPU Hardware

GPU Models H100 SXM A100 SXM L40S RTX 4090 RTX 6000 Ada RTX 5090 RTX 3090
Max VRAM 80 GB
Max GPUs per Instance 8
Interconnect NVLink
Multi-Node Training No

Pricing

Starting Price $0.11/hr
Billing Granularity Per-second
Spot/Preemptible Yes
Reserved Discounts N/A
Free Credits Up to $10,000 for startups
Egress Fees None (Free)
Storage Container disk (60GB free), volume disk, network volumes

On-Demand GPU Instances are billed per-second with hourly settlement. Pricing ranges from $0.21/hr for RTX 3090 (24GB) to $2.59/hr for H100 SXM (80GB). Multi-GPU configurations available up to 8x GPUs per instance.

Spot Instances offer up to 50% off on-demand pricing with a 1-hour guaranteed run time. Spot prices start at $0.11/hr for RTX 3090 and $1.30/hr for H100 SXM.

Storage pricing: Container disk at $0.005/GB/day (60GB free), volume disk at $0.005/GB/day, and network volumes at $0.002/GB/day. Egress is completely free with no bandwidth charges.

Startup Program offers up to $10,000 in free credits for qualifying startups. Reserved plans with 1-12 month commitments provide additional discounts.

Infrastructure

Regions US, EU, APAC, South America, Africa, Middle East (20+ locations)
Uptime SLA 99.9%
Serverless / Autoscaling Yes
Private Networking / VPC No

Developer Experience

Pre-installed Frameworks PyTorch TensorFlow CUDA cuDNN TensorRT
Docker Support Yes
SSH Access Yes
Jupyter Notebooks Yes
API / CLI Yes
Setup Time N/A
Kubernetes Support No
Custom Images / Templates Yes
Persistent Storage Yes

Business Terms

Min Commitment None
Compliance SOC 2
Best For AI training inference fine-tuning generative AI research LLM serving Stable Diffusion
Support Channels Email Discord Documentation
Payment Methods Credit Card
VS

How does it compare?

Compare Novita AI against other cloud GPU providers.

Frequently Asked Questions

Should I use Novita AI for my AI/ML project?

Novita AI caters primarily to users who need: AI training, inference, fine-tuning, generative AI, research, LLM serving, Stable Diffusion

As a GPU-Focused provider, Novita AI differentiates itself through its GPU selection and pricing model. Starting at $0.11/hr, the platform provides accessible compute for researchers, startups, and enterprises building AI applications.

GPU models available: H100 SXM, A100 SXM, L40S, RTX 4090, RTX 6000 Ada, RTX 5090, RTX 3090.

Compare Novita AI features with your project requirements at Novita AI official website.

Is Novita AI well-reviewed on Trustpilot?

The current Trustpilot rating for Novita AI is 2.7 out of 5.0, based on 8 total reviews as of July 14, 2026. Novita AI was founded in 2022.

You can read all user reviews directly on the Trustpilot page for Novita AI. Trustpilot ratings reflect real user experiences with GPU provisioning speed, pricing accuracy, support responsiveness, and overall platform reliability.

See how Novita AI compares to alternatives and explore their current offerings at Novita AI official website.

Can I install my own CUDA toolkit and frameworks on Novita AI?

The ML framework ecosystem at Novita AI:

PyTorch, TensorFlow, CUDA, cuDNN, TensorRT

These frameworks come pre-configured and optimized for the available GPU hardware. Custom images (Yes) allow you to deploy any framework or library not included by default, including nightly builds, custom forks, or specialized inference engines like vLLM or TensorRT.

Persistent storage (Yes) ensures your datasets and model weights survive instance restarts.

Check which ML frameworks come pre-installed on Novita AI official website.

What is the setup and deployment experience like at Novita AI?

Novita AI is designed for developers and researchers who need GPU compute quickly:

- Time to deploy:
- Docker: Yes
- SSH: Yes
- Jupyter: Yes
- API/CLI: Yes
- Custom images: Yes

Docker support allows you to bring your own environment with pre-installed frameworks, CUDA versions, and dependencies. This eliminates environment setup time and ensures reproducibility across development and production.

Launch your first GPU instance in minutes at Novita AI official website.

Does Novita AI offer auto-scaling GPU endpoints?

Serverless GPU at Novita AI: Yes

Serverless GPU inference allows you to deploy models that automatically scale up when requests arrive and scale down to zero when idle, eliminating the cost of keeping GPUs running during quiet periods. This is particularly cost-effective for applications with variable or unpredictable traffic patterns.

Novita AI standard GPU pricing starts at $0.11/hr with Per-second billing.

For serverless GPU endpoint setup guides and pricing, see Novita AI official website.

Does Novita AI offer private networking between GPU instances?

Novita AI operates out of United States with GPU infrastructure available in:

US, EU, APAC, South America, Africa, Middle East (20+ locations)

The platform provides an uptime guarantee of 99.9%. For teams running production-grade inference or training pipelines, the combination of regional availability and SLA commitments should factor into provider selection.

Private networking / VPC: No

For region-specific pricing and compliance details, visit Novita AI official website.

Is NVLink or InfiniBand available at Novita AI?

Here is how Novita AI handles multi-GPU workloads:

GPU interconnect: NVLink
Maximum GPUs: 8 per instance
Multi-node support: No

When scaling beyond a single GPU, the interconnect technology determines how efficiently GPUs communicate during distributed training operations like all-reduce and gradient synchronization. Novita AI offers NVLink connectivity, which is an important factor when comparing multi-GPU providers for large-scale AI workloads.

Check available multi-node cluster configurations on Novita AI official website.

Are there preemptible GPU options at Novita AI for fault-tolerant workloads?

Spot/preemptible instances at Novita AI: Yes

Spot instances offer significantly reduced pricing (typically 50-90% cheaper) in exchange for the possibility that your instance may be interrupted when demand is high. This makes them ideal for fault-tolerant workloads like distributed training with checkpointing, batch inference, and hyperparameter sweeps.

Novita AI standard pricing starts at $0.11/hr with Per-second billing.

Check current spot instance availability and discount rates at Novita AI official website.

Are there hidden bandwidth charges at Novita AI?

Novita AI handles data transfer fees as follows: None (Free)

This matters particularly for teams running distributed training across providers or serving models via API endpoints that return large payloads. Zero or low egress fees can significantly reduce overall costs for production inference deployments.

Available storage: Container disk (60GB free), volume disk, network volumes.

Check the full bandwidth and data transfer pricing on Novita AI official website.

Does Novita AI offer any sign-up bonus or free compute credits?

Free credit availability at Novita AI: Up to $10,000 for startups

Novita AI GPU instances start from $0.11/hr, so any promotional credits can stretch across multiple short training runs or inference tests.

Claim your free GPU credits at Novita AI official website.

What are the GPU specifications available at Novita AI?

As of July 14, 2026, Novita AI provides access to the following GPU models:

H100 SXM, A100 SXM, L40S, RTX 4090, RTX 6000 Ada, RTX 5090, RTX 3090

For memory-intensive workloads such as large language model training, the highest VRAM option at Novita AI is 80 GB. Multi-GPU instances support up to 8 GPUs with NVLink interconnect for efficient parallel computation.

See which GPU models are currently in stock at Novita AI official website.

How does Novita AI charge for GPU compute time?

As of July 14, 2026, the lowest-cost GPU option at Novita AI starts from $0.11/hr. Novita AI uses Per-second billing, so workloads that run for just a few minutes are billed accordingly without paying for an unused full hour.

Reserved pricing:
Spot instances available: Yes

Accepted payment methods include Credit Card.

Compare GPU rental rates and billing options at Novita AI official website.

User Feedback

There are no public trader reviews for this firm yet. If you have traded with them, be the first to leave a short, honest review and help other traders.

Share Your Experience

Short, honest feedback helps other prop traders understand what it is really like to work with this firm.

By sending feedback you agree that your comment can be published on this page. Personal details such as email are never shown publicly.

Security check