Modal sources GPU capacity from Site mentions AWS infrastructure and resells it with a managed layer on top. This model can simplify the buying process for teams that prefer not to deal with the underlying provider directly.
There is a margin built in and transparency over where the hardware runs is limited. Compare against the underlying provider before committing at scale.
Confidence: MediumEvidence reviewed: 3 checks
02 / OVERVIEW
Company overview
Operator class
Reseller
GPU models tracked
A10, A100, L40S and 5 more
Services
GPU compute, Managed services
03 / CAPACITY
GPU lineup by use case
Tracked GPU SKUs grouped by the workload they fit best.
Training
For multi-node runs where interconnect and cluster availability decide the outcome.
A100
QuoteClaim to publish
VRAM 80 GBClass SXM
Best for Mature training stacks that value known performance and broad tooling support.
B200
QuoteClaim to publish
VRAM 192 GBClass Blackwell
Best for Frontier-scale training and teams optimizing around the newest platform.
H100
QuoteClaim to publish
VRAM 80 GBClass SXM
Best for Serious model training where time-to-result outweighs hourly cost.
H200
QuoteClaim to publish
VRAM 141 GBClass Hopper
Best for Large-scale training where memory bandwidth decides the outcome.
Inference
For predictable production serving, latency targets, and steady utilisation.
A10
QuoteClaim to publish
VRAM 24 GBClass Ampere
Best for Inference and rendering for workloads that don't need A10G pricing.
L40S
QuoteClaim to publish
VRAM 48 GBClass Ada
Best for Cost-effective production inference and image generation at scale.
L4
QuoteClaim to publish
VRAM 24 GBClass Ada
Best for Cost-efficient inference for smaller production models and API serving.
T4
QuoteClaim to publish
VRAM 16 GBClass Turing
Best for Cost-optimised inference for smaller models with predictable load.
04 / TRUST
Verification
Evidence that there's a real operator behind the offer.
Indicator
Reseller classification
Modal sources capacity from Site mentions AWS infrastructure and resells access with a managed layer.
Indicator
Provenance
Infrastructure is not independently operated. Buyers requiring direct-operator guarantees should compare against the source provider.
05 / COMMERCIALS
Pricing
Indicative published rates. Final cluster pricing depends on commit length, storage, networking, and capacity.
Monthly rates
$30 – $100 /mo
Across 6 published SKUs spanning A10, A100, L40S, B200.
07 / NEXT
Talk to Modal
viabandwidth does not sit between you and the provider. Go straight to their site.
modal.com
Pricing pages, quote forms, and sales routes live on the provider’s own site.
Profile pages, buyer guides, and model explainers stay open. Vendor packages cover what shows on this profile; verification stays independent and is never for sale.