Spec & Pricing Reference

NVIDIA H100 SXM5 Cloud Pricing & Specs (2026)

The H100 SXM5 is the workhorse of AI training clusters. Its 80 GB HBM3 memory handles 70B parameter models with tensor parallelism across 2-4 GPUs, while 3.35 TB/s bandwidth keeps the compute units fed during all-reduce operations.

Target workload: Multi-node 70B+ LLM pre-training & fine-tuning

Memory80GB HBM3
Bandwidth3.35 TB/s
FP8 TFLOPS1,979
Observed rateNo tracked rows
80 GB VRAM
Manufacturer SpecHIGH
SourceNVIDIA manufacturer specifications
VerifiedSep 30, 2026
Value
80 GB
Methodology

Manufacturer specification for onboard memory.

Refreshed daily from provider APIs and market scrapingSep 30, 2026
3.35 TB/s
Manufacturer SpecHIGH
SourceNVIDIA specification
VerifiedSep 30, 2026
Value
3.35 TB/s GB/s
Methodology

Peak memory bandwidth from the manufacturer specification.

Refreshed daily from provider APIs and market scrapingSep 30, 2026
Methodology →

NVIDIA H100 SXM5: Key Numbers at a Glance

Rental cost: no verified rate rows are tracked yet — this page asserts specifications only, never a price.

70B fit: Fits Llama 3.3 70B at FP8/INT8 (weights ≈71 GB at 1 byte/param) with a short-context KV budget; FP16 (140 GB) requires 2-way tensor parallelism.

Bottleneck: Memory-bandwidth bound at large batch; compute-bound at small batch with FP8/BF16

Specs Only — No Rate Rows Tracked
Pricing coverage for NVIDIA H100 SXM5

OpenGPU Radar does not currently track "H100 SXM5" rows in data/providers.json. No price is asserted on this page — the specifications below are manufacturer-sourced.

Compare GPUs with tracked rates →

Compatible Models for NVIDIA H100 SXM5

Models from the VRAM registry whose minimum INT4 footprint (weights + KV-cache + runtime overhead) fits 80 GB. FP16 shows where full precision also fits single-GPU.

Browse all model VRAM pages →

Specifications

ArchitectureHopper GH100 — 4nm TSMC
Memory80GB HBM3
Bandwidth3.35 TB/s
InterconnectNVLink 4.0 (900 GB/s)
TDP700W
FP16 TFLOPS989
FP8 TFLOPS1,979
FP4 TFLOPSN/A
Recommended quantizationFP8 / FP4 native
Best cluster topology8-way HGX Baseboard with NVLink 4.0 Mesh

H100 SXM5 Memory-Bandwidth Ceiling Analysis: In an 8-way HGX baseboard, each GPU connects via NVLink 4.0 at 900 GB/s bidirectional bandwidth. For Llama 70B FP8 tensor parallelism, TP=2 splits the 72 GB model across 2 GPUs (36 GB each) with 16 GB headroom for KV-cache at 32k context. TP=4 splits across 4 GPUs (18 GB each) but introduces 3 all-reduce sync points per forward pass — the 900 GB/s NVLink mesh absorbs this with <5% communication overhead. At TP=8, the model shards to 9 GB per GPU, but NCCL all-reduce latency increases to ~8% due to the ring-all-reduce topology across 8 nodes. The optimal config is TP=2 for inference (lowest latency) and TP=4 for training (better compute utilization). At batch≥128, the H100's 3.35 TB/s HBM3 bandwidth becomes the limiting factor: the GPU's 1,979 FP8 TFLOPS of compute throughput can only be sustained if the memory subsystem delivers sufficient weight data per cycle. With 80 GB HBM3 at 3.35 TB/s, the effective bandwidth-to-compute ratio is ~1.69 bytes per FLOP, well above the theoretical minimum of ~0.5 bytes/FLOP for FP8 inference, confirming the memory-bandwidth bound.

Next steps

Compare Alternatives

Head-to-head comparisons against this GPU — specs, observed hourly rates, and the workload verdict.

H100 SXM5 vs H200 SXM5

The H200 eliminates 2-GPU tensor sharding for Llama 70B at 128k context. The H100 remains the lower-cost option for sub-32k workloads at $1.89/hr vs $2.50/hr.

H100 SXM5 vs B200 Blackwell

The B200 delivers 2x inference throughput via FP4 but requires liquid-cooled infrastructure at 1000W TDP. The H100 is deployment-ready in air-cooled datacenters today.

H100 SXM5 vs A100 80GB SXM4

H100 SXM5 leads FP8-native serving: 1,979 FP8 TFLOPS and 3.35 TB/s bandwidth deliver a higher decode ceiling per GPU, and NVLink 4.0 at 900 GB/s scales tensor parallelism. The A100 80GB is 16% cheaper per hour observed ($1.59 vs $1.89) and remains the budget choice for INT8/INT4 batch workloads and Ampere-standardized tooling — but it has no FP8 tensor cores, so Hopper's native FP8 pipeline advantage does not exist on Ampere.

H100 SXM5 vs H200 SXM

H100 SXM5 is the cost-efficient workhorse: identical 1,979 FP8 TFLOPS and NVLink 4.0 at $1.89/hr observed — $0.90/hr less than the H200. The H200's 141 GB (+76% VRAM, +43% bandwidth) is the unlock for long-context work: 70B at FP8 with full 128k KV-cache fits one GPU, where the H100's 80 GB forces eviction or shorter context. Choose H100 for cost-efficient standard serving; choose H200 when context length or single-GPU simplicity outruns.

B200 Blackwell vs H100 SXM5

B200 doubles the platform: 192 GB fits 70B at BF16-class precision where the H100 needs FP8, and 8.0 TB/s plus 2,250 FP8 TFLOPS push per-GPU throughput past Hopper. But it costs 2.1× per hour observed ($3.99 vs $1.89). Choose H100 unless you need single-GPU BF16 70B, FP4-capable kernels, or cluster consolidation — the B200's per-GPU advantage only pays when its throughput is actually sustained.

Related GPUs

Frequently Asked Questions

How much does it cost to rent NVIDIA H100 SXM5 per hour?▾
OpenGPU Radar does not currently track rate rows for NVIDIA H100 SXM5. The specifications on this page are manufacturer-sourced; no price is asserted.
Can NVIDIA H100 SXM5 run 70B parameter LLMs?▾
Fits Llama 3.3 70B at FP8/INT8 (weights ≈71 GB at 1 byte/param) with a short-context KV budget; FP16 (140 GB) requires 2-way tensor parallelism.
What is NVIDIA H100 SXM5's memory bandwidth?▾
NVIDIA H100 SXM5 has 3.35 TB/s of memory bandwidth across 80GB HBM3. Token generation is memory-bandwidth bound at batch size 1, so decode throughput scales with this figure — no tokens/sec value is claimed without a benchmark row.
Specifications only — no rate rows tracked for this GPU.Spec sources: NVIDIA manufacturer specifications.Methodology →