Azure Machine Learning
Deploy open or fine-tuned models to managed Azure endpoints, billed per VM-hour.
Overview
Azure Machine Learning lets you deploy open-weight models from its model catalogue, including the Hugging Face collection, or your own fine-tuned models, to managed online endpoints. Endpoints run on GPU virtual machines reserved for you, billed per hour while they run, with savings plans and reserved instances for committed use.
Specifications
| Hosting type | Dedicated endpoint |
|---|---|
| Model families | gpt-oss, Qwen, DeepSeek, Llama, Gemma, Mistral |
| LLM models | gpt-oss-120b, Qwen2.5 72B Instruct, DeepSeek-V3, Llama 3.3 70B Instruct, Llama 3.1 8B Instruct, gpt-oss-20b, Gemma 3 27B, Gemma 3 12B, Gemma 3 4B, Gemma 2 27B, Gemma 2 9B, Qwen3 30B-A3B, Qwen3 32B, Qwen3 235B-A22B, Qwen2.5-Coder 32B Instruct, DeepSeek-V3.1, DeepSeek-R1, Codestral 22B, Mistral Small 3.1, Mistral NeMo |
| Pricing range | From US$0.526 |
| Pricing model | Per GPU-hour, Reserved or committed capacity |
| Free tier available | No |
| Data center locations | United States, Canada, Brazil, Mexico, Ireland, Netherlands, United Kingdom, France, Germany, Switzerland, Norway, Sweden, Poland, Italy, Spain, Israel, Qatar, United Arab Emirates, South Africa, India, Japan, South Korea, Singapore, Hong Kong SAR China, Australia, Indonesia, Malaysia, New Zealand |
| Hardware / GPU options | NCasT4_v3 (NVIDIA T4), NVadsA10 v5 (A10), NC A100 v4 (A100), NC H100 v5 and ND H100 v5 (H100), ND H200 v5 (H200), ND GB200 v6 (GB200) and ND MI300X v5 (AMD MI300X) virtual machines. |
| VRAM & memory | 16 GB – 192 GB |
| Maximum model size | 671B |
| OpenAI-compatible API | No |
News about Microsoft Azure
All news →LLM Hosting Roundup: Anthropic’s capacity crunch, Vera Rubin and DeepSeek’s $15B (23 Oct 2026)
Anthropic's hunt for cloud capacity, NVIDIA's DSX POD and new Vera Rubin inference benchmarks, Dell's first Groq 3 rack, DeepSeek's reported $15B raise and Reflection's open-weights Beam model.
