← All GPUs

Hardware reference / Data Center

NVIDIA H100

Hopper architecture. Device specifications and exact provider configurations, kept separate.

VRAM
80 GB
Unverified record
Memory bandwidth
3,000 GB/s
Unverified record
CUDA cores
14,592
Unverified record
Illustrative NVIDIA H100 hardware image
Catalogue illustration, not proof of a provider's exact board, form factor or available stock.

H100 Provider Offer Comparison

View approved price history

Recorded from $1.99/GPU-hour USD

(runpod: H100 PCIe 80GB; Community Cloud on-demand); confirm rate and stock

28 recorded offers; not live inventory. USD quotes; availability is not live. Compare matching variants and billing units.

Provider Configuration / billing Recorded hourly price GPU count Recorded VRAM Availability Provenance
runpod H100 NVL 94GB; Community Cloud on-demand Recorded $2.59/GPU-hour USD GPU count unknown 94 GB (per GPU) Availability unknown

Verified 2026-10-31

www.runpod.io ↗
Evidence

Price verified against source at 2026-10-31T23:54:57Z; not live inventory. Public advertised GPU-hour quote, not a stock check or total instance cost. GPU count, region, disk/network charges and final deployment terms require provider confirmation.

runpod H100 NVL 94GB; Secure Cloud on-demand Recorded $3.19/GPU-hour USD GPU count unknown 94 GB (per GPU) Availability unknown

Verified 2026-10-31

www.runpod.io ↗
Evidence

Price verified against source at 2026-10-31T23:54:57Z; not live inventory. Public advertised GPU-hour quote, not a stock check or total instance cost. GPU count, region, disk/network charges and final deployment terms require provider confirmation.

runpod H100 PCIe 80GB; Community Cloud on-demand Recorded $1.99/GPU-hour USD GPU count unknown 80 GB (per GPU) Availability unknown

Verified 2026-10-31

www.runpod.io ↗
Evidence

Price verified against source at 2026-10-31T23:54:57Z; not live inventory. Public advertised GPU-hour quote, not a stock check or total instance cost. GPU count, region, disk/network charges and final deployment terms require provider confirmation.

runpod H100 PCIe 80GB; Secure Cloud on-demand Recorded $2.89/GPU-hour USD GPU count unknown 80 GB (per GPU) Availability unknown

Verified 2026-10-31

www.runpod.io ↗
Evidence

Price verified against source at 2026-10-31T23:54:57Z; not live inventory. Public advertised GPU-hour quote, not a stock check or total instance cost. GPU count, region, disk/network charges and final deployment terms require provider confirmation.

runpod H100 SXM 80GB; Community Cloud on-demand Recorded $2.69/GPU-hour USD GPU count unknown 80 GB (per GPU) Availability unknown

Verified 2026-10-31

www.runpod.io ↗
Evidence

Price verified against source at 2026-10-31T23:54:57Z; not live inventory. Public advertised GPU-hour quote, not a stock check or total instance cost. GPU count, region, disk/network charges and final deployment terms require provider confirmation.

runpod H100 SXM 80GB; Secure Cloud on-demand Recorded $3.99/GPU-hour USD GPU count unknown 80 GB (per GPU) Availability unknown

Verified 2026-10-31

www.runpod.io ↗
Evidence

Price verified against source at 2026-10-31T23:54:57Z; not live inventory. Public advertised GPU-hour quote, not a stock check or total instance cost. GPU count, region, disk/network charges and final deployment terms require provider confirmation.

lambda H100 SXM 80GB; On-demand; plus applicable tax Recorded $3.99/GPU-hour USD 8 GPUs 80 GB (per GPU) Availability unknown

Verified 2026-11-01

Source reviewed 2026-11-01

lambda.ai ↗
Evidence

Price verified against source at 2026-11-01T13:40:33.105830+00:00; not live inventory. Public source reviewed 2026-11-01T13:39:29.566Z (not verification of this offer). Explicit agent review by ReviewLambda (OpenAI coding agent; official rendered tabs inspected); manual_browser_import capture 4d14aa2fd9ddd7b6b1d50c5e5cc4adb446d8aca20b98fa20c2e38d5b5bc9df3c; review 2e4555ebcceaff10a8f4565716720a7f6a764b8bde93f80233fca17a09bd2e9c. Advertised source evidence, not live inventory; reviewer attested configuration and units.

lambda H100 SXM 80GB; On-demand; plus applicable tax Recorded $4.09/GPU-hour USD 4 GPUs 80 GB (per GPU) Availability unknown

Verified 2026-11-01

Source reviewed 2026-11-01

lambda.ai ↗
Evidence

Price verified against source at 2026-11-01T13:40:33.114361+00:00; not live inventory. Public source reviewed 2026-11-01T13:39:29.596Z (not verification of this offer). Explicit agent review by ReviewLambda (OpenAI coding agent; official rendered tabs inspected); manual_browser_import capture 5d16ef32c6ea162726ac0190cd0156ff32c99f598b6e1013219ec2989f6aba00; review 757267639999962f1c3d7a29d06adbb0538875cc6014247ee40aede6b924391a. Advertised source evidence, not live inventory; reviewer attested configuration and units.

lambda H100 SXM 80GB; On-demand; plus applicable tax Recorded $4.19/GPU-hour USD 2 GPUs 80 GB (per GPU) Availability unknown

Verified 2026-11-01

Source reviewed 2026-11-01

lambda.ai ↗
Evidence

Price verified against source at 2026-11-01T13:40:33.115514+00:00; not live inventory. Public source reviewed 2026-11-01T13:39:29.629Z (not verification of this offer). Explicit agent review by ReviewLambda (OpenAI coding agent; official rendered tabs inspected); manual_browser_import capture 8014314284939f520b04bbc9ad3f25904f4ffe186379bf0397aa518ef760d143; review b908bbc9b6f6a1ca07960519bc7f897072dda3d18abc15594edc6c8bfef6b0d7. Advertised source evidence, not live inventory; reviewer attested configuration and units.

lambda H100 SXM 80GB; On-demand; plus applicable tax Recorded $4.29/GPU-hour USD 1 GPU 80 GB (per GPU) Availability unknown

Verified 2026-11-01

Source reviewed 2026-11-01

lambda.ai ↗
Evidence

Price verified against source at 2026-11-01T13:40:33.116344+00:00; not live inventory. Public source reviewed 2026-11-01T13:39:29.661Z (not verification of this offer). Explicit agent review by ReviewLambda (OpenAI coding agent; official rendered tabs inspected); manual_browser_import capture 3ead1f3f9601be099997391704c1b8ac6552285e453897efb4d0b92fb3622575; review 469c1ea470bd5e2f7d21ffadea9dc4d71ed9f16cc56de4c0c423104aa607e4fd. Advertised source evidence, not live inventory; reviewer attested configuration and units.

lambda H100 PCIe 80GB; On-demand; plus applicable tax Recorded $3.29/GPU-hour USD 1 GPU 80 GB (per GPU) Availability unknown

Verified 2026-11-01

Source reviewed 2026-11-01

lambda.ai ↗
Evidence

Price verified against source at 2026-11-01T13:40:33.116344+00:00; not live inventory. Public source reviewed 2026-11-01T13:39:29.661Z (not verification of this offer). Explicit agent review by ReviewLambda (OpenAI coding agent; official rendered tabs inspected); manual_browser_import capture 3ead1f3f9601be099997391704c1b8ac6552285e453897efb4d0b92fb3622575; review 469c1ea470bd5e2f7d21ffadea9dc4d71ed9f16cc56de4c0c423104aa607e4fd. Advertised source evidence, not live inventory; reviewer attested configuration and units.

coreweave NVIDIA HGX H100; On-Demand; NORTH AMERICA Recorded $49.24/instance-hour USD ($6.155/GPU-hour) 8 GPUs VRAM unknown Availability unknown

Verified 2026-11-01

Source reviewed 2026-11-01

coreweave.com ↗
Evidence

Price verified against source at 2026-11-01T16:41:47.158Z; not live inventory. Public source reviewed 2026-11-01T16:38:17.245Z (not verification of this offer). Explicit agent review by ReviewCoreWeave (official Chromium browser capture and source-layout review); manual_browser_import capture 93f5460a99dc0622ab5a46745b971a699dd2f3a27b23f40eb1f31cdfd4f00fee; review 27eed2855091f1184c718ffd2f41f21186202bc77a8b0613d11e45a1a4b7dc83. Advertised source evidence, not live inventory; reviewer attested configuration and units.

coreweave NVIDIA HGX H100; Spot; NORTH AMERICA Recorded $19.71/instance-hour USD ($2.464/GPU-hour) 8 GPUs VRAM unknown Availability unknown

Verified 2026-11-01

Source reviewed 2026-11-01

coreweave.com ↗
Evidence

Price verified against source at 2026-11-01T16:41:47.158Z; not live inventory. Public source reviewed 2026-11-01T16:38:17.245Z (not verification of this offer). Explicit agent review by ReviewCoreWeave (official Chromium browser capture and source-layout review); manual_browser_import capture 93f5460a99dc0622ab5a46745b971a699dd2f3a27b23f40eb1f31cdfd4f00fee; review 27eed2855091f1184c718ffd2f41f21186202bc77a8b0613d11e45a1a4b7dc83. Advertised source evidence, not live inventory; reviewer attested configuration and units.

coreweave NVIDIA HGX H100; On-Demand; EUROPE Recorded $49.24/instance-hour USD ($6.155/GPU-hour) 8 GPUs VRAM unknown Availability unknown

Verified 2026-11-01

Source reviewed 2026-11-01

coreweave.com ↗
Evidence

Price verified against source at 2026-11-01T16:41:47.158Z; not live inventory. Public source reviewed 2026-11-01T16:38:17.245Z (not verification of this offer). Explicit agent review by ReviewCoreWeave (official Chromium browser capture and source-layout review); manual_browser_import capture 93f5460a99dc0622ab5a46745b971a699dd2f3a27b23f40eb1f31cdfd4f00fee; review 27eed2855091f1184c718ffd2f41f21186202bc77a8b0613d11e45a1a4b7dc83. Advertised source evidence, not live inventory; reviewer attested configuration and units.

coreweave NVIDIA HGX H100; Spot; EUROPE Recorded $19.51/instance-hour USD ($2.439/GPU-hour) 8 GPUs VRAM unknown Availability unknown

Verified 2026-11-01

Source reviewed 2026-11-01

coreweave.com ↗
Evidence

Price verified against source at 2026-11-01T16:41:47.158Z; not live inventory. Public source reviewed 2026-11-01T16:38:17.245Z (not verification of this offer). Explicit agent review by ReviewCoreWeave (official Chromium browser capture and source-layout review); manual_browser_import capture 93f5460a99dc0622ab5a46745b971a699dd2f3a27b23f40eb1f31cdfd4f00fee; review 27eed2855091f1184c718ffd2f41f21186202bc77a8b0613d11e45a1a4b7dc83. Advertised source evidence, not live inventory; reviewer attested configuration and units.

thundercompute H100 SXM; On-demand Recorded $3.2/GPU-hour USD GPU count unknown VRAM unknown Availability unknown

Verified 2026-11-01

Source reviewed 2026-11-01

www.thundercompute.com ↗
Evidence

Price verified against source at 2026-11-01T16:51:11.819488+00:00; not live inventory. Public source reviewed 2026-11-01T16:38:45.266344+00:00 (not verification of this offer). Explicit agent review by OMP parent agent; official H100 GPU-hour heading and own H100 SXM instance row inspected; manual_browser_import capture 7192c8795a23fa436c2fe21d5e00accd89eea64493ccc6fee702ac371f063174; review b5a0ee05379a4d27bddbdccc3a64f6eb6c806088a1ead516a03ea648018cd6ed. Advertised source evidence, not live inventory; reviewer attested configuration and units.

nebius NVIDIA HGX H100; On-demand Recorded $4.5/GPU-hour USD GPU count unknown VRAM unknown Availability unknown

Verified 2026-11-01

Source reviewed 2026-11-01

nebius.com ↗
Evidence

Price verified against source at 2026-11-01T17:18:58.460416+00:00; not live inventory. Public source reviewed 2026-11-01T17:14:59.434858+00:00 (not verification of this offer). Explicit agent review by OMP agent; official Nebius browser pricing table reviewed; manual_browser_import capture bd90f39838874ccf8f2066e374aaf555fa43383aa71b33b40de9e3884e96ca8e; review 56eee3ffd25f5fdfaba1ba8531723e5df8a81f5bf60a5b4fd620622e1a966889. Advertised source evidence, not live inventory; reviewer attested configuration and units.

jarvislabs H100; billing model unknown Recorded $2.8/hour (unit unverified) USD GPU count unknown 80 GB (scope unverified) Availability unknown

Price unverified

Source unknown

Evidence

Historical catalogue price; verification date unknown

runpod H100; billing model unknown Recorded $2.79/hour (unit unverified) USD GPU count unknown 94 GB (scope unverified) Availability unknown

Price unverified

Source reviewed 2026-10-31

www.runpod.io ↗
Evidence

Historical catalogue price; verification date unknown. Public source reviewed 2026-10-31T23:54:57Z (not verification of this offer). Public page distinguishes H100 PCIe, SXM and NVL with different Community/Secure rates. This generic historical model does not establish a matching configuration or price.

runpod H100; billing model unknown Recorded $2.39/hour (unit unverified) USD GPU count unknown 80 GB (scope unverified) Availability unknown

Price unverified

Source reviewed 2026-10-31

www.runpod.io ↗
Evidence

Historical catalogue price; verification date unknown. Public source reviewed 2026-10-31T23:54:57Z (not verification of this offer). Public page distinguishes H100 PCIe, SXM and NVL with different Community/Secure rates. This generic historical model does not establish a matching configuration or price.

runpod H100; billing model unknown Recorded $2.99/hour (unit unverified) USD GPU count unknown 80 GB (scope unverified) Availability unknown

Price unverified

Source reviewed 2026-10-31

www.runpod.io ↗
Evidence

Historical catalogue price; verification date unknown. Public source reviewed 2026-10-31T23:54:57Z (not verification of this offer). Public page distinguishes H100 PCIe, SXM and NVL with different Community/Secure rates. This generic historical model does not establish a matching configuration or price.

thundercompute H100; billing model unknown Recorded $1.47/hour (unit unverified) USD 1 GPU 80 GB (scope unverified) Recorded availability: Available

Price unverified

Source unknown

Evidence

Historical catalogue price; verification date unknown

thundercompute H100; billing model unknown Recorded $3.29/hour (unit unverified) USD 1 GPU 80 GB (scope unverified) Recorded availability: Coming Soon

Price unverified

Source unknown

Evidence

Historical catalogue price; verification date unknown

hyperbolic H100; billing model unknown Recorded $1.5/hour (unit unverified) USD 1 GPU 80 GB (scope unverified) Recorded availability: Available

Price unverified

Source unknown

Evidence

Historical catalogue price; verification date unknown

verda H100; billing model unknown Recorded $2.49/hour (unit unverified) USD 1 GPU 80 GB (scope unverified) Recorded availability: Available

Price unverified

Source unknown

Evidence

Historical catalogue price; verification date unknown

vultr H100; billing model unknown Recorded $15.92/hour (unit unverified) USD 8 GPUs 640 GB (scope unverified) Recorded availability: Available

Price unverified

Source unknown

Evidence

Historical catalogue price; verification date unknown

tensordock H100; billing model unknown Recorded $2.265/hour (unit unverified) USD 1 GPU 80 GB (scope unverified) Recorded availability: available

Price unverified

Source unknown

Evidence

Historical catalogue price; verification date unknown

digitalocean H100; billing model unknown Recorded $5.95/hour (unit unverified) USD 1 GPU 80 GB (scope unverified) Recorded availability: available

Price unverified

Source unknown

Evidence

Historical catalogue price; verification date unknown

Specifications & reference

Recorded specifications describe the device, not necessarily every provider configuration. Check source scope before comparing.

Recorded Technical Specifications
CUDA cores
14,592
Base clock
1,410 MHz
Boost clock
1,830 MHz
Memory
80 GB; HBM3
Memory bus
5,120 bit
Recorded TDP / board power
700 W
Recorded Performance
Recorded FP32
Unknown
Recorded FP16
Unknown

Only source-verified benchmark fields are shown. Match precision, sparsity, form factor and test conditions before comparison; unknown means no verified measurement is recorded.

Detailed Specifications
Architecture Hopper
GPU code name Unknown
Recorded release date 2022-03-22
Recorded launch price (not rental) 30,000 USD
Manufacturing process 4nm
Transistors 80B

Specification review: not recorded · 0 source-verified fields · Cloud availability: unknown

Specification sources & scope

Only the listed fields are source-verified; other specifications remain unverified. Source review does not establish live availability.

Source reviewed
No source review recorded
Source-verified fields
None recorded; historical specifications remain unverified
Specification scope
Unknown; exact form factor and configuration require confirmation
Lifecycle evidence
Unknown
Lifecycle note
No lifecycle detail recorded
Cloud rental availability
Unknown — catalogue inclusion is not confirmation of rental stock

Manufacturer source links

No specification source URL recorded.

Recorded AI features

Only source-verified fields are shown; unknown does not mean unsupported. Confirm hardware and software versions.

Tensor Cores

Unknown

Transformer Engine

Unknown

Flash Attention

Unknown

Physical specifications

Physical Specifications

Dimensions

10.5in

Length

4.4in

Width

2-slot

Height

Software & workload notes

Allow memory for weights, optimizer state, KV cache and framework overhead. Core counts and nominal throughput are not directly comparable across architectures or manufacturers.

Recorded CUDA compute capability: 9.0. CUDA is NVIDIA-specific; confirm driver, toolkit and framework requirements against NVIDIA's documentation.

Manufacturer & software references

Confirm support for the exact device, operating system and software version.

NVIDIA CUDA documentation ↗NVIDIA GPU compute capability reference ↗
Partner listings · affiliate links

H100 Partner Listings

We may earn a commission from these links. A recorded listing does not guarantee stock or current pricing.

Thunder Compute

Recorded from $3.2/GPU-hour USD (thundercompute: H100 SXM; On-demand); confirm rate and stock

3 exact-model recorded offers; confirm price and availability.

Visit Thunder Compute →

RunPod

Recorded from $1.99/GPU-hour USD (runpod: H100 PCIe 80GB; Community Cloud on-demand); confirm rate and stock

9 exact-model recorded offers; confirm price and availability.

Visit RunPod →