Documentation – OpenMetal Cloud

The RTX PRO 6000’s 96GB VRAM fits 70B models at FP8 on a single card with real KV cache headroom. This article covers what that unlocks, how dedicated fixed-cost GPU infrastructure compares structurally to cloud rental, and where the H200 is the better choice.

NVIDIA H200 vs H100 for AI training and inference: 141GB HBM3e vs 80–94GB, same Hopper compute with more memory. OpenMetal runs the H200 on bare metal.

NVIDIA RTX Pro 6000 vs H200 on OpenMetal: 96GB GDDR7 + FP4 for cost-efficient AI vs 141GB HBM3e for the largest models. Both single-tenant bare metal.

OpenMetal NVIDIA H200 bare metal GPU server: 141GB HBM3e, dual Xeon 6530P, 1TB DDR5. Single-tenant bare metal, fixed monthly pricing.

OpenMetal GPU clusters: dedicated single-tenant multi-GPU infrastructure. All-RP6000, all-H200, or mixed on a private 40 Gbps mesh, fixed monthly pricing.

OpenMetal NVIDIA RTX Pro 6000 GPU server: 96GB GDDR7, FP4, dual Xeon 6530P, 1TB DDR5. Training and inference, single-tenant, fixed monthly pricing.

Q: What is the difference between the NVIDIA RTX Pro 6000 and H100? The RTX Pro 6000 is a Blackwell GPU with 96GB of GDDR7 and native FP4, while the

Q: Is the RTX Pro 6000 better than the L40S for AI inference and training? For most training and inference the RTX Pro 6000 outperforms the L40S on a single

Add NVIDIA RTX Pro 6000 or H200 GPU servers to an existing OpenMetal cloud or bare metal deployment – same private network, fixed monthly pricing.

NVIDIA RTX Pro 6000 vs H100: specs, cost, deployment fit. 96GB GDDR7 + FP4 vs 80–94GB HBM3. OpenMetal offers the RP6000 and H200 on bare metal.

NVIDIA RTX Pro 6000 vs L40S for AI training and inference: 96GB GDDR7 + FP4 (Blackwell) vs 48GB GDDR6 (Ada). OpenMetal runs the RP6000 on bare metal.

Q: Can I run Intel TDX confidential computing on an OpenMetal GPU server? Intel TDX and GPU passthrough cannot be combined in a single trust boundary on OpenMetal GPU servers,

Q: Can I attach RP6000 GPU nodes to an existing OpenMetal bare metal or Hosted Private Cloud deployment? Yes, you can attach RP6000 GPU nodes to an existing OpenMetal Hosted

Q: What is the difference between the NVIDIA RTX Pro 6000 and L40S? The RTX Pro 6000 is a newer Blackwell-generation GPU with 96GB of GDDR7 and native FP4, while

Q: Can I build a mixed GPU cluster with RP6000 and H200 servers? Yes, OpenMetal builds mixed GPU clusters that combine RP6000 and H200 nodes on the same private network,

Q: What is FP4 (NVFP4) and why does it matter for AI workloads? FP4 (NVFP4) is a Blackwell-native 4-bit floating-point format that increases low-precision inference throughput beyond the FP8 ceiling

Q: How does OpenMetal’s fixed-cost GPU pricing avoid the cloud “idle silicon tax”? OpenMetal charges a fixed monthly rate for a dedicated GPU server, so running the card at 100%

Q: Can I train and fine-tune AI models on the OpenMetal RP6000, or is it only for inference? Yes, the OpenMetal RP6000 trains and fine-tunes AI models as well as

Q: How much GPU memory does the OpenMetal RP6000 have? Each OpenMetal RP6000 GPU carries 96GB of GDDR7 memory, and a server can hold one or two cards for up

Q: GDDR7 vs HBM3: which matters for AI training and inference? GDDR7 offers high capacity at lower cost, while HBM3/HBM3e delivers much higher memory bandwidth; bandwidth is what matters most

Q: Can I run a 70B parameter LLM on a single OpenMetal H200? Yes, a single OpenMetal H200 runs a 70B-parameter model in 16-bit precision, because its 141GB of HBM3e

Q: Can I build a multi-GPU cluster with OpenMetal H200 servers? Yes, OpenMetal builds dedicated multi-GPU clusters of H200 servers on a private 40 Gbps mesh, built to order for

Q: Can I add GPU servers to my existing OpenMetal cloud or bare metal deployment? Yes, you can add NVIDIA RTX Pro 6000 or H200 GPU servers to an existing

Q: What NVMe storage does the OpenMetal H200 GPU server use? The OpenMetal H200 GPU server uses a 6.4TB Micron 7500 MAX NVMe SSD for data, plus two 960GB NVMe

Q: What CPU is paired with the OpenMetal H200 GPU server? Each OpenMetal H200 GPU server pairs the GPU with two Intel Xeon 6530P processors (Granite Rapids), giving 64 cores

Q: Should I choose the RP6000 or the H200 for my workload? Choose the RP6000 for cost-efficient training, fine-tuning, and high-throughput inference that fit in 96GB, and the H200 when

Q: How does OpenMetal GPU pricing compare to AWS GPU instances? OpenMetal prices GPU servers on a fixed monthly model with included egress, while AWS bills GPU instances per GPU-hour

Q: What is the difference between the NVIDIA H200 and H100? The H200 and H100 share the same Hopper compute architecture; the H200’s advantage is memory, with 141GB of HBM3e

Q: Is the NVIDIA H200 faster than the H100 for AI inference? For memory-bound LLM inference, yes: the H200’s higher HBM3e bandwidth (4.8 TB/s vs 3.35-3.9 TB/s) directly raises tokens-per-second,

Q: Why does OpenMetal offer the NVIDIA H200 instead of the H100? OpenMetal carries the H200 rather than the H100 because the H200 is the H100’s direct successor: 50% more

The control plane fee is the smallest part of your managed Kubernetes bill. This article breaks down what EKS, GKE, and AKS actually charge across egress, storage, cross-zone transfer, and multi-cluster overhead, and where self-managed on dedicated bare metal makes the math work better.

DORA has been in force since January 2025, and the third-party ICT risk requirements are where infrastructure decisions land hardest. This article breaks down what Articles 28–30 require, why hyperscaler concentration is now a documented regulatory problem, and how private cloud in the EU changes the risk picture.

Egress is the infrastructure cost most teams don’t model until it’s already on the bill. This article explains how per-GB and 95th percentile billing models work, why your 95th percentile figure isn’t your average usage, and how OpenMetal’s included allocation plus flat overage rate compares.

Learn how to enable Intel SGX and TDX on OpenMetal’s v4 and v5 servers. This guide covers required memory configurations (full channel allotment and 1TB RAM), hardware prerequisites, and a detailed cost comparison for provisioning SGX/TDX-ready infrastructure.

The v5 generation can be told as a cores-and-clocks story, but a significant change is bandwidth: the private fabric doubled to 40 Gbps, memory moved to DDR5-6400, and the lane budget grew to 88 PCIe 5.0 lanes.

Running AI inference on sensitive data requires hardware-level isolation, not just software controls. This guide covers how to build a confidential inference pipeline on OpenMetal’s XL v5 using Intel TDX, including Trust Domain setup, vLLM deployment, attestation, and storage architecture.

All-NVMe OSDs, an isolated boot pool, a clean lane budget, and identical nodes: how OpenMetal’s v5 hardware makes Ceph behave predictably instead of needing tuning.

The OpenMetal XL v5 is built on dual Intel Xeon 6530P processors (Granite Rapids, Intel 3 process) with 1TB DDR5-6400, 25.6TB of Micron 7500 MAX NVMe, and full Intel TDX support as a base configuration. This article covers the workloads it’s built for, why TDX matters for specific use cases, how the private cloud and bare metal configurations compare, and where it fits in the v5 lineup relative to the Large.

OpenMetal Large v4 Hosted Private Cloud: 3-node OpenStack + Ceph cluster with 96 cores, 1.5TB DDR5, 38.4TB NVMe. Deploy in 45 seconds, fixed monthly pricing, no VMware licensing

The Hosted Private Cloud Medium v4 is a three-node OpenStack and Ceph cluster built on the same Medium v4 hardware available as a standalone bare metal server. Each node contributes

The Hosted Private Cloud Medium v5 is a three-node OpenStack and Ceph cluster built on the same Medium v5 hardware available as a standalone bare metal server. Each node contributes

The Medium v4 TDX Edition is not a separate server model. It is the standard Medium v4 chassis with all 16 DIMM slots fully populated — 8 × 64 GB

The OpenMetal Bare Metal Dedicated Server XL v4 TDX Edition is not a separate server model — it is the XL v4 in its standard 1TB RAM configuration, with Intel

Q: Does the OpenMetal XL v4 support Intel TDX confidential computing? The OpenMetal XL v4 ships with Intel TDX active by default — no RAM upgrade needed; OpenMetal enables TDX

The OpenMetal Hosted Private Cloud XL v4 is a three-node OpenStack and Ceph cluster, each node running dual Intel Xeon Gold 6530 processors with 1TB DDR5 4800MHz RAM and 25.6TB

The OpenMetal Hosted Private Cloud on XXL v4 hardware delivers a three-node OpenStack + Ceph cluster built on the highest-density compute and storage nodes in the v4 generation — ready

This page covers the OpenMetal XXL v4 configured as a confidential computing platform. The XXL v4 is the only server in the OpenMetal v4 lineup where Intel TDX (Trust Domain

Every scaling company eventually reaches an infrastructure inflection point. Explore the five stages of infrastructure maturity, the economics of cloud repatriation, hybrid cloud strategy, and how infrastructure ownership can improve cost predictability, control, and long-term flexibility.

Modern cloud platforms reward speed, and the abstractions that make systems easy to build also shape how they are built. This article examines how architectural convenience accumulates into dependency, why lock-in is structural rather than contractual, and what intentional friction has to do with sovereignty.

Q: What is the difference between Intel TDX and Intel SGX on OpenMetal XL v5? Intel TDX isolates an entire guest VM (a “trust domain”) from the host hypervisor and