Runpod vs Lambda vs Vast.ai: which GPU cloud fits your workload
Runpod's pods and serverless, Lambda's instances and clusters, Vast.ai's host marketplace: how billing, reliability and commitment differ.
Choose Runpod if…
- You want one account for dev pods, a serverless inference endpoint and later a multi-node cluster
- Per-second billing with no commitment matters, and you want a choice between Runpod-operated Secure Cloud and cheaper Community Cloud
- You need consumer or workstation GPUs (RTX 4090, 5090) alongside data-centre GPUs
Choose Lambda if…
- You want NVIDIA data-centre instances (1 to 8 GPUs) with local NVMe, run by a single provider
- You plan a multi-node run: 1-Click Clusters start at 16 GPUs for two weeks with published per-GPU rates
- You can work from Lambda's region list (US, Japan, India, Germany, Israel)
Choose Vast.ai if…
- Lowest hourly price matters more than consistency, and your job tolerates an occasional host problem
- You can filter by host reliability score and verification status yourself
- You want a wide range of GPU models and locations from many independent hosts
| Provider | GPU | Offering | USD / GPU-hour | Billing | Commitment | Regions | Source |
|---|---|---|---|---|---|---|---|
| Vast.ai | NVIDIA B200192 GB | MarketplaceOn-demand marketplace listing (1 GPU) | $7.50 lowest*listed: lowest listing $7.502/hrmedian $10.01 across 2 listings | per second | None stated | USEUAPACOTHER | Vast.ai pricing (live offers API)8 Nov 2026 |
| Runpod | NVIDIA B200192 GB | On-demandSecure Cloud pod | $7.99listed: $7.99/hr | per second | None stated | UNSPECIFIED | Runpod pricing8 Nov 2026 |
| Lambda | NVIDIA B200192 GBmin 16× | Reserved1-Click Cluster, 16 GPUs | $9.86listed: $9.86 per GPU/hr | not-stated | 2 weeks to 1 year term, 16 GPUs | USEUAPACOTHER | Lambda pricing8 Nov 2026 |
| Lambda | NVIDIA B200192 GB | On-demand1x B200 SXM6 | Not publishedAnnounced $7.99 from 16 Nov 2026 (not yet in effect) | not-stated | None stated | USEUAPACOTHER | Lambda pricing8 Nov 2026 |
| Lambda | NVIDIA B200192 GBmin 8× | On-demand8x HGX B200 | Not publishedAnnounced $7.69 from 16 Nov 2026 (not yet in effect) | not-stated | None stated | USEUAPACOTHER | Lambda pricing8 Nov 2026 |
| Vast.ai | NVIDIA H200141 GB | MarketplaceOn-demand marketplace listing (1 GPU) | $3.23 lowest*listed: lowest listing $3.231/hrmedian $4.34 across 8 listings | per second | None stated | USEUAPACOTHER | Vast.ai pricing (live offers API)8 Nov 2026 |
| Runpod | NVIDIA H200141 GB | CommunityCommunity Cloud pod | from $3.59listed: from $3.59/hr | per second | None stated | UNSPECIFIED | Runpod pricing8 Nov 2026 |
| Runpod | NVIDIA H200141 GB | On-demandSecure Cloud pod | $5.29listed: $5.29/hr | per second | None stated | UNSPECIFIED | Runpod pricing8 Nov 2026 |
| Vast.ai | NVIDIA H100 SXM80 GB | MarketplaceOn-demand marketplace listing (1 GPU) | $2.28 lowest*listed: lowest listing $2.278/hrmedian $2.67 across 5 listings | per second | None stated | USEUAPACOTHER | Vast.ai pricing (live offers API)8 Nov 2026 |
| Runpod | NVIDIA H100 SXM80 GB | CommunityCommunity Cloud pod | from $2.69listed: from $2.69/hr | per second | None stated | UNSPECIFIED | Runpod pricing8 Nov 2026 |
| Runpod | NVIDIA H100 SXM80 GB | On-demandSecure Cloud pod | $3.99listed: $3.99/hr | per second | None stated | UNSPECIFIED | Runpod pricing8 Nov 2026 |
| Lambda | NVIDIA H100 SXM80 GBmin 16× | Reserved1-Click Cluster, 16 GPUs | $6.16listed: $6.16 per GPU/hr | not-stated | 2 weeks to 1 year term, 16 GPUs | USEUAPACOTHER | Lambda pricing8 Nov 2026 |
| Lambda | NVIDIA H100 SXM80 GB | On-demand1x H100 SXM | Not publishedAnnounced $4.89 from 16 Nov 2026 (not yet in effect) | not-stated | None stated | USEUAPACOTHER | Lambda pricing8 Nov 2026 |
| Lambda | NVIDIA H100 SXM80 GBmin 8× | On-demand8x HGX H100 | Not publishedAnnounced $4.59 from 16 Nov 2026 (not yet in effect) | not-stated | None stated | USEUAPACOTHER | Lambda pricing8 Nov 2026 |
| Vast.ai | NVIDIA A100 80GB80 GB | MarketplaceOn-demand marketplace listing (1 GPU) | $0.63 lowest*listed: lowest listing $0.630/hrmedian $1.14 across 8 listings | per second | None stated | USEUAPACOTHER | Vast.ai pricing (live offers API)8 Nov 2026 |
| Runpod | NVIDIA A100 80GB80 GB | CommunityCommunity Cloud pod | from $1.39listed: from $1.39/hr | per second | None stated | UNSPECIFIED | Runpod pricing8 Nov 2026 |
| Runpod | NVIDIA A100 80GB80 GB | On-demandSecure Cloud pod | $1.79listed: $1.79/hr | per second | None stated | UNSPECIFIED | Runpod pricing8 Nov 2026 |
| Lambda | NVIDIA A100 80GB80 GBmin 8× | On-demand8x HGX A100 80GB | Not publishedAnnounced $3.19 from 16 Nov 2026 (not yet in effect) | not-stated | None stated | USEUAPACOTHER | Lambda pricing8 Nov 2026 |
Normalised to US dollars per GPU per hour from each provider's published price; "listed" shows the wording as published. "Contact sales" and "Not published" mean no public price was found. We never estimate one. Where a provider prices a multi-GPU instance, the per-GPU figure is the instance price divided by its GPU count. * Marketplace prices are set by individual hosts and change constantly; the figure is the lowest live listing on the verification date, with the median of the listings sampled. Announced prices that have not yet taken effect are shown separately and never as the current price. Prices exclude storage, egress and taxes unless the row says otherwise.
All three are self-serve: you can sign up and start a GPU without a sales call. They differ in who operates the hardware and how you are billed, which matters more than a few cents per hour.
How they are built
- Runpod runs two pools. Secure Cloud runs in data centres Runpod operates, Community Cloud on third-party hosts. It also sells Serverless workers and multi-node Clusters. Pods are “billed per second with no commitment”.
- Lambda rents NVIDIA instances with 1, 2, 4 or 8 GPUs, with local NVMe included, plus 1-Click Clusters. Its docs list 14 regions across the US, Japan, India, Germany and Israel; not every instance type is in every region.
- Vast.ai is a marketplace. Independent hosts list machines and set prices (“Prices are set by the market, not by Vast”), with on-demand and interruptible listings and per-second billing. Reliability varies by host, and listings carry a reliability score.
Prices: read with care
- Runpod publishes Secure Cloud prices per GPU-hour and “from” prices for Community Cloud.
- Lambda. As of 8 November 2026, Lambda’s current on-demand prices are visible only to logged-in users. The prices on lambda.com take effect on 16 November 2026, and the index shows them as announced, not current. Its cluster prices are current.
- Vast.ai has no single list price. The index shows the lowest and median single-GPU on-demand listings in its live offers feed at the time we checked, and those change continuously.
So a straight “cheapest of the three” answer would mix a list price, a future price and a market snapshot. Use the table to see the range, and the cheapest GPU cloud by GPU class page for list-price comparisons only.
Commitment and scale
- If you need more than one node, Lambda publishes per-GPU cluster rates (two weeks to one year, 16 GPUs or more), while Runpod’s H100 and B200 clusters are “Contact sales” (its H200 SXM and A100 SXM clusters are listed).
- Vast.ai sells reserved 1, 3 or 6-month terms through sales.
Other GPU clouds
The GPU Cloud Index also covers CoreWeave, Nebius, Hyperstack, DigitalOcean, Modal and Baseten.
| Provider | GPU | Offering | USD / GPU-hour | Regions | Source |
|---|---|---|---|---|---|
| Vast.ai | NVIDIA L40S48 GB | MarketplaceOn-demand marketplace listing (1 GPU) | $0.535 lowest*listed: lowest listing $0.535/hrmedian $0.801 across 9 listings | USEUAPACOTHER | Vast.ai pricing (live offers API)8 Nov 2026 |
| Runpod | NVIDIA L40S48 GB | On-demandSecure Cloud pod | $1.09listed: $1.09/hr | UNSPECIFIED | Runpod pricing8 Nov 2026 |
| Vast.ai | NVIDIA GeForce RTX 509032 GB | MarketplaceOn-demand marketplace listing (1 GPU) | $0.40 lowest*listed: lowest listing $0.400/hrmedian $0.592 across 64 listings | USEUAPACOTHER | Vast.ai pricing (live offers API)8 Nov 2026 |
| Runpod | NVIDIA GeForce RTX 509032 GB | On-demandSecure Cloud pod | $1.19listed: $1.19/hr | UNSPECIFIED | Runpod pricing8 Nov 2026 |
| Vast.ai | NVIDIA GeForce RTX 409024 GB | MarketplaceOn-demand marketplace listing (1 GPU) | $0.135 lowest*listed: lowest listing $0.135/hrmedian $0.456 across 64 listings | USEUAPACOTHER | Vast.ai pricing (live offers API)8 Nov 2026 |
| Runpod | NVIDIA GeForce RTX 409024 GB | On-demandSecure Cloud pod | $0.89listed: $0.89/hr | UNSPECIFIED | Runpod pricing8 Nov 2026 |
Normalised to US dollars per GPU per hour from each provider's published price; "listed" shows the wording as published. "Contact sales" and "Not published" mean no public price was found. We never estimate one. * Marketplace prices are set by individual hosts and change constantly; the figure is the lowest live listing on the verification date, with the median of the listings sampled. Prices exclude storage, egress and taxes unless the row says otherwise.
Provider facts
Runpod
GPU pods on Secure Cloud (Runpod-operated data centres) or Community Cloud (vetted third-party hosts), plus Serverless endpoints and multi-node Clusters.
- Billing
- per second
- Self-serve
- Yes
- Regions
- UNSPECIFIED
- Commitment
- None for Pods ("billed per second with no commitment"); reserved capacity via sales
- Storage
- Volume disk $0.10/GB/mo running, $0.20/GB/mo idle; network storage from $0.05/GB/mo
- Footprint
- "30+ regions" (no region list on the pricing page)
Verified 8 Nov 2026 · Runpod pricing
Lambda
On-demand GPU instances (1, 2, 4 or 8 GPUs) and 1-Click Clusters of 16+ GPUs with a two-week minimum term.
- Billing
- not-stated
- Self-serve
- Yes
- Regions
- US, EU, APAC, OTHER
- Current on-demand prices
- Shown to logged-in users only until a new public list takes effect on 16 Nov 2026
- Regions
- 14 listed in docs: US (8), Japan (2), India, Germany, Israel; not every type in every region
- Storage
- Local NVMe included with each instance type
Verified 8 Nov 2026 · Lambda pricing, Lambda docs: on-demand instances
Vast.ai
A marketplace: independent hosts set prices for on-demand and interruptible GPU rentals; Vast.ai also sells reserved terms through sales.
- Billing
- per second
- Self-serve
- Yes
- Regions
- US, EU, APAC, OTHER
- Pricing
- "Prices are set by the market, not by Vast"; per-second billing, no minimum
- Reliability
- Varies by host; listings show a reliability score and verification status
- Reserved
- 1, 3 or 6-month terms via sales ("up to 50% off")
Verified 8 Nov 2026 · Vast.ai pricing (live offers API)
Sources
- Runpod pricing, checked 8 Nov 2026
- Runpod docs: Serverless pricing, checked 8 Nov 2026
- Lambda pricing, checked 8 Nov 2026
- Lambda docs: on-demand instances (regions), checked 8 Nov 2026
- Vast.ai pricing and public offers API, checked 8 Nov 2026
Index rows carry their own source link and verification date. This page is revisited when: any of the three changes prices, billing granularity, rental models or minimum terms; Lambda's announced list takes effect on 16 Nov 2026