aliteq.

H100 NVL 94GB, what it runs and what it costs to rent

Specs checked · Rental prices from · Aliteq Team · How we compute this

Some links on this page are referral links: we may earn a commission, at no cost to you.

The short answer

The NVIDIA H100 NVL has 94 GB of VRAM and 3900 GB/s of memory bandwidth (2023). At Q4_K_M quality or better with an 8K context it runs 50 of the open models we track; the largest that fits with room to spare is Qwen3.5 122B-A10B (MoE). The cheapest rental right now is $0.667/hr (Vast.ai, interruptible).

VRAM
94 GB
Bandwidth
3900 GB/s
Board power
400 W
Launched
2023

Source: NVIDIA specification · bandwidth not yet confirmed by a second source, so we don't estimate speed from it

Rent the H100 NVL

ProviderTypeFrom, per hourRent
Vast.aiSpot (can be interrupted)$0.667/hrVast.ai
RunpodOn demand$3.190/hrRunpod
Vast.aiOn demand$3.388/hrVast.ai

Lowest price per provider in our hourly snapshot (). Rent buttons are referral links. Every card side by side: cloud GPU prices.

Before you rent the H100 NVL: cheapest H100 rental · Runpod vs Vast.ai vs Lambda · is Vast.ai safe · is Runpod worth it · rent or buy · what an H100 costs to buy · every AI GPU's buy and rent price.

H100 NVL rental price history

The cheapest H100 NVL offer we saw each day on the providers we track, since Aug 20. Renting it on demand costs $3.19/hr now; the lowest since Nov 2 was $1.60/hr on Nov 13 (Vast.ai), so today is 99% above that.

On demand now
$3.19/hr
Lowest on demand since Nov 2
$1.60/hr · Nov 13
Cheapest offer now
$0.667/hr
Lowest offer since Aug 20
$0.533/hr · Nov 13
  • Cheapest offer, incl. interruptible
  • Cheapest on demand (from Nov 2)
$0$2$4Aug 20Oct 4nowlowest offer $0.533/hrAug 20: cheapest $1.33/hr on Vast.aiAug 21: cheapest $1.75/hr on Vast.aiAug 22: cheapest $2.34/hr on Vast.aiAug 23: cheapest $2.27/hr on Vast.aiAug 24: cheapest $1.33/hr on Vast.aiAug 25: cheapest $1.33/hr on Vast.aiAug 26: cheapest $1.47/hr on Vast.aiAug 27: cheapest $1.82/hr on Vast.aiAug 28: cheapest $1.73/hr on Vast.aiAug 29: cheapest $2.34/hr on Vast.aiAug 30: cheapest $2.34/hr on Vast.aiAug 31: cheapest $2.34/hr on Vast.aiSep 1: cheapest $2.59/hr on RunpodSep 2: cheapest $2.40/hr on Vast.aiSep 3: cheapest $1.54/hr on Vast.aiSep 4: cheapest $1.54/hr on Vast.aiSep 5: cheapest $1.54/hr on Vast.aiSep 6: cheapest $1.54/hr on Vast.aiSep 7: cheapest $1.54/hr on Vast.aiSep 8: cheapest $1.54/hr on Vast.aiSep 9: cheapest $1.54/hr on Vast.aiSep 10: cheapest $1.54/hr on Vast.aiSep 11: cheapest $2.00/hr on Vast.aiSep 12: cheapest $2.00/hr on Vast.aiSep 13: cheapest $1.96/hr on Vast.aiSep 14: cheapest $2.08/hr on Vast.aiSep 15: cheapest $2.19/hr on Vast.aiSep 16: cheapest $2.27/hr on Vast.aiSep 17: cheapest $2.27/hr on Vast.aiSep 18: cheapest $2.59/hr on RunpodSep 19: cheapest $2.14/hr on Vast.aiSep 20: cheapest $2.14/hr on Vast.aiSep 21: cheapest $2.59/hr on RunpodSep 22: cheapest $2.59/hr on RunpodSep 23: cheapest $1.82/hr on Vast.aiSep 24: cheapest $2.59/hr on RunpodSep 25: cheapest $2.27/hr on Vast.aiSep 26: cheapest $2.21/hr on Vast.aiSep 27: cheapest $2.27/hr on Vast.aiSep 28: cheapest $2.27/hr on Vast.aiSep 29: cheapest $2.14/hr on Vast.aiSep 30: cheapest $2.32/hr on Vast.aiOct 1: cheapest $2.14/hr on Vast.aiOct 2: cheapest $2.14/hr on Vast.aiOct 3: cheapest $2.44/hr on Vast.aiOct 4: cheapest $2.29/hr on Vast.aiOct 5: cheapest $1.99/hr on Vast.aiOct 6: cheapest $1.93/hr on Vast.aiOct 7: cheapest $2.37/hr on Vast.aiOct 8: cheapest $2.25/hr on Vast.aiOct 9: cheapest $2.34/hr on Vast.aiOct 10: cheapest $2.39/hr on Vast.aiOct 11: cheapest $2.13/hr on Vast.aiOct 12: cheapest $1.82/hr on Vast.aiOct 13: cheapest $2.31/hr on Vast.aiOct 14: cheapest $2.19/hr on Vast.aiOct 15: cheapest $1.95/hr on Vast.aiOct 16: cheapest $1.64/hr on Vast.aiOct 17: cheapest $1.35/hr on Vast.aiOct 18: cheapest $1.34/hr on Vast.aiOct 19: cheapest $2.07/hr on Vast.aiOct 20: cheapest $2.31/hr on Vast.aiOct 21: cheapest $2.34/hr on Vast.aiOct 22: cheapest $2.23/hr on Vast.aiOct 23: cheapest $2.31/hr on Vast.aiOct 24: cheapest $2.33/hr on Vast.aiOct 25: cheapest $2.33/hr on Vast.aiOct 26: cheapest $2.33/hr on Vast.aiOct 27: cheapest $2.33/hr on Vast.aiOct 28: cheapest $2.40/hr on Vast.aiOct 29: cheapest $2.47/hr on Vast.aiOct 30: cheapest $2.34/hr on Vast.aiOct 31: cheapest $2.34/hr on Vast.aiNov 1: cheapest $0.667/hr on Vast.aiNov 2: cheapest $0.667/hr on Vast.ai, on demand $2.34/hr on Vast.aiNov 3: cheapest $0.667/hr on Vast.ai, on demand $2.34/hr on Vast.aiNov 4: cheapest $0.667/hr on Vast.ai, on demand $2.34/hr on Vast.aiNov 5: cheapest $0.667/hr on Vast.ai, on demand $2.71/hr on Vast.aiNov 6: cheapest $0.667/hr on Vast.ai, on demand $2.34/hr on Vast.aiNov 7: cheapest $0.667/hr on Vast.ai, on demand $2.34/hr on Vast.aiNov 8: cheapest $0.667/hr on Vast.ai, on demand $2.53/hr on Vast.aiNov 9: cheapest $0.667/hr on Vast.ai, on demand $2.34/hr on Vast.aiNov 10: cheapest $0.667/hr on Vast.ai, on demand $2.34/hr on Vast.aiNov 11: cheapest $0.667/hr on Vast.ai, on demand $2.34/hr on Vast.aiNov 12: cheapest $0.667/hr on Vast.ai, on demand $2.34/hr on Vast.aiNov 13: cheapest $0.533/hr on Vast.ai, on demand $1.60/hr on Vast.aiNov 14: cheapest $0.667/hr on Vast.ai, on demand $2.34/hr on Vast.aiNov 15: cheapest $0.667/hr on Vast.ai, on demand $2.34/hr on Vast.aiNov 16: cheapest $0.533/hr on Vast.ai, on demand $2.34/hr on Vast.aiNov 17: cheapest $0.667/hr on Vast.ai, on demand $2.81/hr on Vast.aiNov 18 (now): cheapest $0.667/hr on Vast.ai, on demand $3.19/hr on Runpod

Daily minimum of our hourly readings of Vast.ai, Runpod and Lium. Interruptible (spot) offers can be taken back mid-job. Before Nov 2 Vast mixed spot into on-demand and Runpod listed Community Cloud rates, so the on-demand line starts then. Every number is in our open daily price file.

Every day as a table
DayCheapest offerCheapest on demand
Nov 18 (now)$0.667/hr · Vast.ai$3.19/hr · Runpod
Nov 17$0.667/hr · Vast.ai$2.81/hr · Vast.ai
Nov 16$0.533/hr · Vast.ai$2.34/hr · Vast.ai
Nov 15$0.667/hr · Vast.ai$2.34/hr · Vast.ai
Nov 14$0.667/hr · Vast.ai$2.34/hr · Vast.ai
Nov 13$0.533/hr · Vast.ai$1.60/hr · Vast.ai
Nov 12$0.667/hr · Vast.ai$2.34/hr · Vast.ai
Nov 11$0.667/hr · Vast.ai$2.34/hr · Vast.ai
Nov 10$0.667/hr · Vast.ai$2.34/hr · Vast.ai
Nov 9$0.667/hr · Vast.ai$2.34/hr · Vast.ai
Nov 8$0.667/hr · Vast.ai$2.53/hr · Vast.ai
Nov 7$0.667/hr · Vast.ai$2.34/hr · Vast.ai
Nov 6$0.667/hr · Vast.ai$2.34/hr · Vast.ai
Nov 5$0.667/hr · Vast.ai$2.71/hr · Vast.ai
Nov 4$0.667/hr · Vast.ai$2.34/hr · Vast.ai
Nov 3$0.667/hr · Vast.ai$2.34/hr · Vast.ai
Nov 2$0.667/hr · Vast.ai$2.34/hr · Vast.ai
Nov 1$0.667/hr · Vast.ai—
Oct 31$2.34/hr · Vast.ai—
Oct 30$2.34/hr · Vast.ai—
Oct 29$2.47/hr · Vast.ai—
Oct 28$2.40/hr · Vast.ai—
Oct 27$2.33/hr · Vast.ai—
Oct 26$2.33/hr · Vast.ai—
Oct 25$2.33/hr · Vast.ai—
Oct 24$2.33/hr · Vast.ai—
Oct 23$2.31/hr · Vast.ai—
Oct 22$2.23/hr · Vast.ai—
Oct 21$2.34/hr · Vast.ai—
Oct 20$2.31/hr · Vast.ai—
Oct 19$2.07/hr · Vast.ai—
Oct 18$1.34/hr · Vast.ai—
Oct 17$1.35/hr · Vast.ai—
Oct 16$1.64/hr · Vast.ai—
Oct 15$1.95/hr · Vast.ai—
Oct 14$2.19/hr · Vast.ai—
Oct 13$2.31/hr · Vast.ai—
Oct 12$1.82/hr · Vast.ai—
Oct 11$2.13/hr · Vast.ai—
Oct 10$2.39/hr · Vast.ai—
Oct 9$2.34/hr · Vast.ai—
Oct 8$2.25/hr · Vast.ai—
Oct 7$2.37/hr · Vast.ai—
Oct 6$1.93/hr · Vast.ai—
Oct 5$1.99/hr · Vast.ai—
Oct 4$2.29/hr · Vast.ai—
Oct 3$2.44/hr · Vast.ai—
Oct 2$2.14/hr · Vast.ai—
Oct 1$2.14/hr · Vast.ai—
Sep 30$2.32/hr · Vast.ai—
Sep 29$2.14/hr · Vast.ai—
Sep 28$2.27/hr · Vast.ai—
Sep 27$2.27/hr · Vast.ai—
Sep 26$2.21/hr · Vast.ai—
Sep 25$2.27/hr · Vast.ai—
Sep 24$2.59/hr · Runpod—
Sep 23$1.82/hr · Vast.ai—
Sep 22$2.59/hr · Runpod—
Sep 21$2.59/hr · Runpod—
Sep 20$2.14/hr · Vast.ai—
Sep 19$2.14/hr · Vast.ai—
Sep 18$2.59/hr · Runpod—
Sep 17$2.27/hr · Vast.ai—
Sep 16$2.27/hr · Vast.ai—
Sep 15$2.19/hr · Vast.ai—
Sep 14$2.08/hr · Vast.ai—
Sep 13$1.96/hr · Vast.ai—
Sep 12$2.00/hr · Vast.ai—
Sep 11$2.00/hr · Vast.ai—
Sep 10$1.54/hr · Vast.ai—
Sep 9$1.54/hr · Vast.ai—
Sep 8$1.54/hr · Vast.ai—
Sep 7$1.54/hr · Vast.ai—
Sep 6$1.54/hr · Vast.ai—
Sep 5$1.54/hr · Vast.ai—
Sep 4$1.54/hr · Vast.ai—
Sep 3$1.54/hr · Vast.ai—
Sep 2$2.40/hr · Vast.ai—
Sep 1$2.59/hr · Runpod—
Aug 31$2.34/hr · Vast.ai—
Aug 30$2.34/hr · Vast.ai—
Aug 29$2.34/hr · Vast.ai—
Aug 28$1.73/hr · Vast.ai—
Aug 27$1.82/hr · Vast.ai—
Aug 26$1.47/hr · Vast.ai—
Aug 25$1.33/hr · Vast.ai—
Aug 24$1.33/hr · Vast.ai—
Aug 23$2.27/hr · Vast.ai—
Aug 22$2.34/hr · Vast.ai—
Aug 21$1.75/hr · Vast.ai—
Aug 20$1.33/hr · Vast.ai—

Which AI models run on the H100 NVL

Computed by our VRAM engine for an 8K context: the best quality each model fits at in 94 GB. Speeds are upper-bound estimates from memory bandwidth, not benchmarks.

ModelBest fitVRAM used~tok/s
Qwen3.5 122B-A10B (MoE)Q5_K_M84 GB89%—
NVIDIA Nemotron 3.5 Super VL 120B-A12B (MoE)Q5_K_M83 GB88%—
NVIDIA Nemotron 3 Super 120B-A12B (MoE)Q5_K_M83 GB88%—
Mistral Small 4 119B (MoE)Q6_K92 GB98% · tight—
gpt-oss 120B (MoE)MXFP462 GB66%—
Llama 4 Scout 17B-16E (MoE)Q6_K84 GB89%—
GLM-4.5-Air (MoE)Q6_K83 GB88%—
Qwen3-Next 80B-A3B (MoE)Q8_080 GB85%—
Qwen3-Coder-Next 80B-A3B (MoE)Q8_080 GB85%—
Llama 3.3 70B InstructQ8_073 GB78%—
DeepSeek R1 Distill 70BQ8_073 GB78%—
Mixtral 8x7B (MoE)F1689 GB94% · tight—

Fits only below Q4 quality: Qwen3 235B-A22B (MoE) (Q2_K), MiniMax-M2.5 (MoE) (Q2_K), Qwen3.8-Flash-Next (MoE) (IQ4_XS).

Too big for 94 GB (rent a bigger card instead): what it costs to run MiniMax-M3 (MoE), what it costs to run MiniMax-M3.1-Flash (MoE), what it costs to run GLM-5.3 (MoE), what it costs to run DeepSeek V4.1 Flash (MoE).

Check a model and context of your own in the VRAM calculator.

AI video on the H100 NVL

Alibaba's Wan 2.2 is an open video model you can run yourself in ComfyUI. With 94 GB the H100 NVL fits:

  • Wan 2.2 14B at 720p without offloading (80 GB and up, per Runpod's Wan 2.2 guide)
  • Wan 2.2 14B in fp16, or in fp8 at 720p (48 GB and up, per Runpod's Wan 2.2 guide)
  • Wan 2.2 14B at 640x640 with fp8 weights (24 GB and up, per Runpod's Wan 2.2 guide)
  • Wan 2.2 5B, with ComfyUI's native offloading (8 GB and up, per ComfyUI's Wan 2.2 docs)

What each setup needs, the cheapest card to rent and what one clip costs: cloud GPUs for ComfyUI with Flux and Wan 2.2. Against an app or an API: AI video cost per clip, app vs API vs rented GPU. Which open model to run, and the licence catch for US users: the best open source AI video generator. Picking by job instead (YouTube, TikTok, ads, animation)? See the best AI video generator for each job. Rather not run a model yourself? Hosted video models such as Seedance, Veo and Hailuo are compared in Kling AI alternatives, priced per clip.

Our guides on the H100 NVL

Cards to compare it with

Every card by VRAM tier: all GPUs for gaming and local AI.

Common questions about the H100 NVL

How much VRAM does the H100 NVL have?
94 GB, with 3900 GB/s of memory bandwidth, per the maker's specification.
What is the biggest AI model the H100 NVL can run?
Of the open models we track, Qwen3.5 122B-A10B (MoE) is the largest that fits at Q4_K_M or better with an 8K context and room to spare (about 84 GB of 94 GB). Bigger models only fit at lower-quality quants or need more memory.
How much does it cost to rent an H100 NVL?
From $0.667/hr on Vast.ai (spot, can be interrupted), the lowest price in our hourly snapshot of Nov 18, 2026. Prices move every hour.
Is 94 GB enough?
Datacenter GPUs and unified-memory machines; mostly rented by the hour.