Rent a B200 at the new training silicon ceiling. 192 GB HBM3e at 8,000 GB/s, Blackwell FP4 tensor cores, 5th-gen NVLink at 1.8 TB/s. Pretraining 405B dense models hits ~14,000 tok/s/GPU on 8-GPU pods, DeepSeek-V3 671B MoE serves at ~3,800 tok/s aggregated, Llama-3 70B FP4 at ~5,200 tok/s with 128-request concurrency. Frontier-lab default. Billed per-minute, paid in BTC, USDT/USDC or CLORE.
B200s are what you rent when minutes matter. FP8 cuts memory and doubles throughput on Transformer workloads; HBM3e keeps the SMs fed; NVLink lets eight cards train as one.
192 GB per card and 8 TB/s HBM3e fit 671B MoE active weights with NVLink-Switch fabric — the only silicon where trillion-parameter-class models serve in production today. 5th-gen NVLink at 1.8 TB/s and FP4 tensor cores roughly double H100 pretraining throughput.
TensorRT-LLM, vLLM, SGLang — all tuned for Blackwell. Serve 70B FP8 from a single card with margin to spare for KV-cache.
192 GB HBM3e + FlashAttention-3 means 128k-token contexts run without offload. Ideal for long-doc RAG and agent loops with tool use.
Blackwell is the architecture the 70B and 405B foundation models were trained on. Specs from Nvidia's SXM5 datasheet; pricing reflects the lowest live on-demand floor.
// prices are spot-market lows · refreshed every 60 s
Every server is priced by its host. These are the live floors across the marketplace — you'll see hundreds of variants once you're in.
No sales call. No quota request. No three-week procurement. The first four commands are all you need.
Filter the marketplace by B200 80GB, country, GPU count, reliability score, network speed.
Choose a Docker image — PyTorch, vLLM, ComfyUI, Blender — or paste your own.
You get a public endpoint, an SSH key, and Jupyter on port 8888 in under 90 s.
Per-minute billing rounds to the second. Stop the instance and the meter stops with it.
Yes — CLORE.AI hosts have started listing B200 nodes as supply ramps in 2026. Availability varies by region; filter by 'B200' in the marketplace and check live spot floor. NVLink-Switch fabric available on multi-node pods.
Single-card, no - 70B pretraining needs an 8-GPU node minimum. CLORE.AI lists 8x H100, 8x H200, and 8x B200 pods with NVLink fabric for exactly this. A100 80GB pods run 70B FSDP training but at lower throughput than Hopper-class. For multi-week training, contact host operators for reserved-instance terms - listed in the marketplace under 'Reserved'.
A100 80GB has no FP8 - peak is BF16/TF32. H100 introduces FP8 with TransformerEngine and roughly 4x the BF16 training throughput at 2x the rental price - so ~2x perf-per-dollar on FP8-eligible workloads. H200 matches H100 compute but adds 141 GB HBM3e. B200 doubles H100 FP8 again with 192 GB HBM3e. Pick by VRAM and bandwidth ceiling, not just sticker FLOPS.
8-GPU H100 SXM, H200 SXM, and B200 nodes ship with NVSwitch fabric - 900 GB/s peer bandwidth on H100/H200, 1.8 TB/s 5th-gen NVLink on B200. PCIe variants (H100 PCIe, A100 PCIe) have NVLink Bridge in pairs only. Multi-node fabric (NVLink-Switch across racks) is available on B200 hyperscale pods - filter by 'NVSwitch' in the marketplace.
Yes. Multi-GPU listings expose all cards in a single rental as a coherent node with NVSwitch (where present), shared NVMe scratch, and InfiniBand or 100 GbE fabric for multi-node training. The standard PyTorch torchrun, DeepSpeed, and Megatron-LM launchers run unmodified. Filter the marketplace by GPU count to find 8x A100, 8x H100, 8x H200 nodes.
V100 (HBM2, 900 GB/s) -> A100 40GB (HBM2e, 1,555 GB/s) -> A100 80GB (HBM2e, 1,935 GB/s) -> H100 (HBM3, 3,350 GB/s) -> H200 (HBM3e, 4,800 GB/s) -> B200 (HBM3e, 8,000 GB/s). Each generation roughly doubles bandwidth or VRAM; KV-cache-bound serving and bandwidth-bound training scale almost linearly with this number.
192 GB HBM3e + 8 TB/s + Blackwell FP4 — the 2026 frontier card for 405B pretraining and 671B MoE serving.
5th-gen NVLink at 1.8 TB/s and FP4 tensor cores roughly double H100 pretraining throughput — frontier-lab default.
Read the guide →192 GB per card holds 671B MoE active weights with NVLink-Switch fabric — the production card for trillion-class MoE.
Read the guide →Native FP4 on Blackwell roughly doubles H100 FP8 70B serving throughput at much higher concurrency.
Read the guide →Side-by-side specs across the datacenter tier. Click any row to see that GPU.
Step-by-step guides verified on CLORE.AI hardware. Pick a workload, copy the docker image, ship in minutes.
Per-minute payouts in BTC, USDT, USDC or CLORE. No listing fee, no contracts, withdraw any time.
Hosts around the world are accepting workloads right now. Sign up, top up your wallet, and the next hour is yours.