Novita AI Model APIs & GPU Cloud
Inference API provider
Provided by Novita AI
Offering overview
Open-weight models can be called without running any servers through a single OpenAI-style endpoint on Novita, which lists more than 200 text, image, audio, video and vision models billed per token or per media unit. Dedicated endpoints with isolated compute are available for teams that need predictable throughput. Beyond the model catalogue, the platform rents GPU instances you control, offers job-based serverless GPUs that scale to zero, provides per-second agent sandboxes for tool-running agents, and sells bare-metal clusters built on 8x H100 and 8x H200 nodes. Current catalogue highlights include DeepSeek V4 Pro, MiniMax M2.7, GLM 5.1, Kimi K2.6, Gemma 4 31B and Qwen3.5 397B. Pricing posture is aggressive: Llama 3.1 8B Instruct is listed at $0.02 per million input tokens, several small models are free, and the company markets rates up to 50% below major clouds. Governing law is Delaware; no HQ address is published.
- Management
- Not stated
- Infrastructure
- Not stated
- Pricing model
- Not stated
- Support
- Not stated
- SLA
- Not stated
- Best for
- Model Inference, AI Agents, AI Applications, Model Training, Fine-Tuning, Batch Processing
Technical specifications
- Containers
- Unknown
- Kubernetes
- Unknown
- Root access
- Unknown
- API access
- Unknown
- Autoscaling
- Unknown
- Persistent processes
- Unknown
AI capabilities
- Model serving
- DeepSeek V4 Pro, MiniMax M2.7, GLM 5.1, Kimi K2.6, Gemma 4 31B, Qwen3.5 397B A17B, Llama 3.1 8B Instruct
- AI frameworks
- Not stated
- Inference endpoints
- Unknown
- Fine-tuning
- Unknown
- Vector database
- Unknown
- Model catalog
- Unknown
- Monitoring
- Not stated
- Evaluation
- Not stated
Security & compliance
No controls are stated.
Geography
Provider headquarters are separate from offering infrastructure and availability.
Infrastructure regions
Not stated
Data-residency jurisdictions
Not stated
Service availability
Not stated
Support coverage
Not stated
Available plans
Pricing can change. Promotional prices are shown separately from standard prices.
| Plan | Price | Compute | Storage | GPU | Price checked | Actions |
|---|---|---|---|---|---|---|
| Novita AI Model APIs & GPU CloudNot stated | $0.02Usage Only | Unknown — RAM | Unknown | No | 7 October 2026 |
Questions & answers
- What models can teams access through Novita?
- Novita lists more than 200 text, image, audio, video and vision models through a single OpenAI-style endpoint, without requiring users to run servers.
- What GPU deployment options does Novita offer?
- Options include user-controlled GPU instances, job-based serverless GPUs that scale to zero, and bare-metal clusters built on 8x H100 and 8x H200 nodes.
- Does Novita support dedicated model endpoints?
- Yes. Dedicated endpoints provide isolated compute for teams that need predictable throughput.
- What are Novita's agent sandboxes for?
- Novita provides sandboxes billed per second for agents that run tools.
Data sources & provenance
Where this information came from, how it was checked and when it was last looked at. “Last updated” records a change to the record — it does not mean the record was verified.
- Data quality
- Data quality: Unverified
- Primary source
- novita.ai
- Sources cited
- 2
- Last reviewed
- 7 October 2026
- Last updated
- 18 October 2026