Offering overview
Two hosting paths sit on top of the Hugging Face Hub. Inference Endpoints provisions a dedicated, autoscaling deployment of any Hub model (or a custom container) on AWS, Azure or Google Cloud, with vLLM, TGI, TEI, SGLang and llama.cpp as serving engines and Kubernetes, CUDA and networking handled for you. Inference Providers takes the opposite approach: a unified API that routes requests to partner clouds such as Groq, Cerebras, Together, Fireworks, fal and Replicate, so developers can call DeepSeek, Llama, Qwen or FLUX models without managing infrastructure. Endpoint hardware ranges from Intel Sapphire Rapids CPUs through NVIDIA T4, L4, A10G, L40S, A100, H100, H200 and B200 GPUs, plus AWS Inferentia2 and Google TPU v5e. Billing is per minute at published hourly rates; CPU instances start at $0.033/hr, the cheapest GPU is a single T4 at $0.50/hr, and an A100 80GB costs $2.50/hr. Routed Inference Providers calls are charged at the partner's own rates, with small monthly credits for paid accounts. Hub users who want production serving without leaving the ecosystem are the audience, from free-tier hobbyists to Enterprise Hub organizations needing SSO and spending controls.
- Management
- Not stated
- Infrastructure
- Not stated
- Pricing model
- Not stated
- Support
- Not stated
- SLA
- Not stated
- Best for
- Model Inference, AI Applications, Fine-Tuning, Model Training, Research And Experimentation, Batch Processing
Technical specifications
- Containers
- Unknown
- Kubernetes
- Unknown
- Root access
- Unknown
- API access
- Unknown
- Autoscaling
- Unknown
- Persistent processes
- Unknown
AI capabilities
- Model serving
- DeepSeek-V3, Llama 3.x, Qwen, GPT-OSS, FLUX.1, BERT, GPT-2
- 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 |
|---|---|---|---|---|---|---|
| Hugging Face Inference Endpoints & Inference ProvidersNot stated | $0.50Usage Only | Unknown — RAM | Unknown | No | 7 October 2026 |
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
- huggingface.co
- Sources cited
- 2
- Last reviewed
- 7 October 2026
- Last updated
- 7 October 2026