Dedicated GPUs
Dedicated NVIDIA and AMD GPUs billed per minute for GPU VMs, fine-tuning runs and autoscaling production inference endpoints.
Find alternatives to Dedicated GPUs
What it does
- GPU cloud
- Model inference
Sources
- Pricing
-
Pricing page Dedicated tab. Page text fetched 2026-11-16 with curl; page quote: "Per GPU-hour, on-demand* B200 $6.25 /hr H200 $4.45 /hr H100 $3.50 /hr A100 $1.80 /hr".
- Description
-
Official documentation · 16 Nov 2026
Own product page, listed in the site nav. Page text fetched 2026-11-16 with curl; page quote: "Dedicated NVIDIA and AMD GPUs for GPU VMs, fine-tuning and production inference at scale.".
- What it does
-
Official documentation · 16 Nov 2026
Own product page, listed in the site nav. Page text fetched 2026-11-16 with curl; page quote: "Dedicated NVIDIA and AMD GPUs for GPU VMs, fine-tuning and production inference at scale.".
- Sold within
-
Published on-demand per-GPU-hour rate card for this product. Page text fetched 2026-11-16 with curl; page quote: "Per GPU-hour, on-demand* B200 $6.25 /hr H200 $4.45 /hr H100 $3.50 /hr A100 $1.80 /hr".
- Name
-
Official documentation · 16 Nov 2026
Own product page, listed in the site nav. Page text fetched 2026-11-16 with curl; page quote: "Dedicated NVIDIA and AMD GPUs for GPU VMs, fine-tuning and production inference at scale.".
- URL
-
Official documentation · 16 Nov 2026
Own product page, listed in the site nav. Page text fetched 2026-11-16 with curl; page quote: "Dedicated NVIDIA and AMD GPUs for GPU VMs, fine-tuning and production inference at scale.".
- Deployment
-
Page describes FlexAI-hosted fleets; saas is the vendor-hosted reading.
Also from FlexAI
-
Token Factory Model API
Token Factory is FlexAI's OpenAI-compatible API to open text, vision, code and image models, billed per token.
-
AgentOS Agent platform
Runs a team's existing agents, skills and tools on open models in FlexAI's harness with approval gates and fully audited traces.
-
AI Factory Infrastructure service
FlexAI's model APIs, AgentOS, dedicated GPUs and an audit log deployed on the customer's own hardware, from VPC to air-gapped, across NVIDIA, AMD and Tenstorrent clusters.
Something wrong here? Send a correction — quote this product id: flexai-dedicated-gpus.