Lantern WorksRent RTX 4090: price today
Cheapest on-demand among tracked providers
$0.67/GPU-hr · Novita AI · RTX 4090 · On-demand.
Cheapest spot / community among tracked providers
$0.16/GPU-hr · SaladCloud · RTX 4090 · Lowest (Batch) priority, interruptible. May be preempted.
Some links are referral links; Lantern Works may earn credit. Affiliate status is not a ranking factor.
On-demand
| Vendor | Variant | VRAM | Tier | $/GPU-hr | Since 2026-10-28 | Egress / storage | Vendor quote | Verified | Link |
|---|---|---|---|---|---|---|---|---|---|
| Novita AI | RTX 4090 | 24 GB | On-demand | $0.67 | newly tracked | egress: not published / not verified; storage: not published / not verified | “RTX 4090 On-Demand The specific price is subject to the computing…”“RTX 4090 On-Demand The specific price is subject to the computing power market. $0.67/GPU/hr” | 2026-10-29 17:00 UTC (13:00 ET) | Referral link |
| RunPod | RTX 4090 | 24 GB | On-demand (Secure Cloud) | $0.89 | no change | egress: not published / not verified; storage: Container disk $0.10/GB/mo; volume disk $0.10/GB/mo running, $0.20/GB/mo idle; network storage $0.07/GB/mo under 1 TB, $0.05/GB/mo over 1 TB. | “On-demand RTX 4090 cloud GPU rental on Runpod. Community Cloud…”“On-demand RTX 4090 cloud GPU rental on Runpod. Community Cloud $0.34/hr, Secure Cloud $0.89/hr.” | 2026-10-29 17:00 UTC (13:00 ET) | Referral link |
Spot & community
| Vendor | Variant | VRAM | Tier | $/GPU-hr | Since 2026-10-28 | Egress / storage | Vendor quote | Verified | Availability (manual snapshot) | Link |
|---|---|---|---|---|---|---|---|---|---|---|
| SaladCloud | RTX 4090 | 24 GB | Lowest (Batch) priority, interruptible | $0.16 · may be preempted | newly tracked | egress: not published / not verified; storage: not published / not verified | “GPU VRAM High Medium Low Lowest $/mo (Lowest) RTX 5090 32 GB $0.600…”“GPU VRAM High Medium Low Lowest $/mo (Lowest) RTX 5090 32 GB $0.600 $0.483 $0.367 $0.250 $182.50 RTX 4090 24 GB $0.440 $0.347 $0.253 $0.160” | 2026-10-29 17:00 UTC (13:00 ET) | — | Direct link |
| RunPod | RTX 4090 | 24 GB | On-demand (Community Cloud) | $0.34 | no change | egress: not published / not verified; storage: Container disk $0.10/GB/mo; volume disk $0.10/GB/mo running, $0.20/GB/mo idle; network storage $0.07/GB/mo under 1 TB, $0.05/GB/mo over 1 TB. | “On-demand RTX 4090 cloud GPU rental on Runpod. Community Cloud…”“On-demand RTX 4090 cloud GPU rental on Runpod. Community Cloud $0.34/hr, Secure Cloud $0.89/hr.” | 2026-10-29 17:00 UTC (13:00 ET) | — | Referral link |
| Vast.ai | RTX 4090 | 24 GB | Marketplace on-demand (verified; median per GPU) | $0.46 (median of 46 offers; lowest $0.3781) | down -10.9% since 2026-10-28 | egress: not published / not verified; storage: not published / not verified | API fields: Vast public offers API: 46 verified rentable on-demand RTX 4090 24GB offers; per-GPU dph_total median $0.4610/hr, min $0.3781/hr | 2026-10-29 17:01 UTC (13:01 ET) | available when checked, 47 offers, regions AU/BG/CA/CN/DK/EE…, snapshot 2026-10-29 15:19 UTC (separate later check) | Referral link |
| Clore.ai | RTX 4090 | 24 GB | Marketplace on-demand (host USD listings; median per GPU) | $0.56 (median of 109 offers; lowest $0.3229) | newly tracked | egress: not published / not verified; storage: not published / not verified | API fields: Clore.ai API: 109 available RTX 4090 24GB host USD listings, range $0.3229-$2.0833; per-GPU median $0.555/hr, min $0.3229/hr | 2026-10-29 15:27 UTC (11:27 ET) | — | Referral link |
Monthly
| Vendor | Variant | VRAM | Tier | $/GPU-hr | Since 2026-10-28 | Egress / storage | Vendor quote | Verified | Link |
|---|---|---|---|---|---|---|---|---|---|
| Ucartz | RTX 4090 | 24 GB | Dedicated server (monthly) | $798.72/mo monthly flat | no change | egress: Traffic allowance is included per plan (amount varies by plan; see the plan table).; storage: NVMe storage included per plan (each plan lists its exact disks). | “RTX 4090 servers from $798.72” | 2026-10-29 17:00 UTC (13:00 ET) | Referral link |
7-day change appears once price history spans 7 days (history starts 2026-10-27).
TensorDock not shown: held until re-verified (vendor pricing page last updated 2024-07-24).
Read before comparing
Rankings are the cheapest among tracked providers on our check date, not across the whole market. Vendors we don't track may cost less.
GPU-hours vs wall-clock time
Prices are per GPU-hour: one GPU for one hour. Four GPUs for 10 wall-clock hours is 40 GPU-hours. Per-second billing (serverless) charges only while your code runs; hourly pods bill from start to stop, idle time included; monthly servers bill a flat price whatever you use. A faster GPU can finish the same job in fewer GPU-hours, so the lowest hourly rate isn't always the lowest cost per job.
Hardware differences
The same GPU name can mean different hardware. SXM modules link GPUs over NVLink and run at higher power limits than PCIe cards, so an H100 SXM is usually faster than an H100 PCIe, especially for multi-GPU training. Interconnect between GPUs and nodes (NVLink, InfiniBand, plain Ethernet) is rarely listed on pricing pages; treat it as unknown unless the vendor states it. VRAM decides which models fit at all. Hardware as listed: RTX 4090 24 GB (not stated).
Egress and storage extras
Totals on this page exclude these. Each cell cites the vendor page it came from; "Not published" means the vendor doesn't publish it there.
| Vendor | Egress | Storage | Interruption / billing terms |
|---|---|---|---|
| Clore.ai | Not published: Not verified in this pass. (checked 2026-10-29) | Not published: Not verified in this pass. (checked 2026-10-29) | Not published: Not verified in this pass. (checked 2026-10-29) |
| Novita AI | Not published: Not published on the pricing page. (checked 2026-10-29) | Not published: Not published on the pricing page. (checked 2026-10-29) | On-Demand price; the vendor says it is subject to the computing power market. Spot pricing is not listed here (unreconciled 'from' headline). (source, 2026-10-29) |
| RunPod | Not published: Not published on the pricing page. (checked 2026-10-27) | Container disk $0.10/GB/mo; volume disk $0.10/GB/mo running, $0.20/GB/mo idle; network storage $0.07/GB/mo under 1 TB, $0.05/GB/mo over 1 TB. (source, 2026-10-27) | Listed rates are on-demand Pod prices (Secure Cloud and Community Cloud). (source, 2026-10-27) |
| SaladCloud | Not published: Not published on the pricing page. (checked 2026-10-29) | Not published: Not published on the pricing page. (checked 2026-10-29) | Four priority tiers. Only the Lowest (Batch) tier is listed here: its price is unchanged by the price change taking effect 2026-11-01 00:00 UTC, while High/Medium/Low rise on that date. Lowest is 'the lowest cost and the most interruptible'; every tier runs on distributed nodes subject to node disconnection. (source, 2026-10-29) |
| Ucartz | Traffic allowance is included per plan (amount varies by plan; see the plan table). (source, 2026-10-27) | NVMe storage included per plan (each plan lists its exact disks). (source, 2026-10-27) | Dedicated servers billed monthly with no setup fee. (source, 2026-10-27) |
| Vast.ai | Not published: Not verified in this pass. (checked 2026-10-27) | Not published: Not verified in this pass. (checked 2026-10-27) | Marketplace on-demand offers (median of live API offers). Interruptible tier is cheaper and not included. (source, 2026-10-27) |
Interruptible vs on-demand
Interruptible (preemptible or spot) capacity can be stopped by the vendor mid-job and is usually cheaper; on-demand runs until you stop it. Interruptible (preemptible) rates listed here: SaladCloud RTX 4090. All other rates are on-demand: the vendor doesn't publish preemption for them.
FAQ
What is the cheapest RTX 4090 among tracked providers per hour?
Among tracked providers, Novita AI at $0.67 per GPU-hour on-demand (On-demand), verified 2026-10-29.
Is spot cheaper for RTX 4090?
The cheapest spot/community rate among tracked providers is $0.16 (SaladCloud) vs $0.67 on-demand. Spot or interruptible capacity may be preempted.
Are egress and storage included?
No. Totals exclude storage, egress, extra vCPU and taxes; each vendor's published terms (or 'not published') are shown in the table.
How often are prices checked?
A daily job re-reads every vendor pricing page; a rate not re-verified for 7 days is held and never ranked.
More
Other GPUs: A100 · H100 · H200 · L40S · RTX 5090 · RTX A6000
GPU price hub · price changes · Data source: gpu-prices.json · gpu-price-changes.json (RSS) · history.json