Loading prices…
RTX 4070 SUPER rental prices
NVIDIA · Ada Lovelace · 12 GB GDDR6X · PCIe · 2024
1 providers · 12 configurations · prices as of Nov 11, 2026, 07:02 UTC
Other versions:12GB
Lowest on-demand
$0.09per GPU·h · ≈ $69/month
at Vast.aiMedian on-demand
$0.09/GPU·h
across 1 providers
Lowest spot
$0.12/GPU·h
interruptible
FP16/BF16 cost
$1.34/PFLOP·h
#6 of 75 GPUs for value
Providers
1
12 configurations
Price history
Daily RTX 4070 SUPER prices per GPU-hour. The shaded period is reconstructed from public archives: the version history of SkyPilot's price catalogs, Internet Archive copies of providers' pricing pages and the open gpu-price-feed project (fewer providers than live data). After that, PetaPrice's own tracking.
Lowest on-demandMedian on-demandLowest spot
Price by provider
Cheapest On-demand RTX 4070 SUPER offer at each provider, per GPU-hour. Shorter and greener is cheaper.
Vast.ai🇺🇸$0.09best
Thin vertical line = median across providers ($0.09).
Every provider we track
Pricing
Size
Host
Provider HQ
1 providers · 12 configurations
Sort
- Vast.aicommunityOn-demandSpot🇺🇸 HQ in United Statesvia Vast.aifrom · On-demandCheapest$0.09/GPU·hCheapest
- 6× GPU · PCIe80 vCPU · 94 GB RAM · 3,342 GB disk$0.09 /GPU·h$0.57 /hOn-demand🇨🇳 Anhui, CN● availableRent ↗
- 4× GPU · PCIe40 vCPU · 94 GB RAM · 955 GB disk$0.10 /GPU·h$0.38 /hOn-demand🇺🇸 United States, US● availableRent ↗
- 2× GPU · PCIe20 vCPU · 47 GB RAM · 477 GB disk$0.10 /GPU·h$0.19 /hOn-demand🇺🇸 United States, US● availableRent ↗
What can the RTX 4070 SUPER run?
Memory for the weights plus one 32K-token sequence of context, how many GPUs hold it, and what that replica costs at the cheapest in-stock on-demand rate.
| Model | Params | Memory | GPUs | $/hr | $/mo |
|---|---|---|---|---|---|
gpt-oss-20b OpenAI · 20.9B | 20.9B · 3.6B active | 20 GiB | 2 | $0.19 at Vast.ai | $139 |
Gemma 3 27B Google · 27.4B | 27.4B | 28 GiB | 4 | $0.38 at Vast.ai | $277 |
Qwen3-32B Alibaba · 32.8B | 32.8B | 39 GiB | 4 | $0.38 at Vast.ai | $277 |
Llama 3.3 70B Instruct Meta · 70.6B | 70.6B | 76 GiB | 8 | $0.76 at Vast.ai | $554 |
gpt-oss-120b OpenAI · 117B | 117B · 5.1B active | 110 GiB | 16 | $1.52 at Vast.ai | $1,108 |
Qwen3-235B-A22B Alibaba · 235B | 235B · 22B active | 225 GiB | 32 | $3.04 at Vast.ai | $2,217 |
DeepSeek-R1 DeepSeek · 684B | 684B · 37B active | 640 GiB | 64 | $6.07 at Vast.ai | $4,434 |
Kimi K3 Moonshot AI · 2.78T | 2.78T · 104B active | 2.53 TiB | 64+ | — | — |
Assumes 90% of each GPU's memory is usable and rounds up to 1, 2, 4, 8… GPUs for tensor parallelism (up to 64). $/hr is the GPU count times the cheapest in-stock on-demand rate per GPU; check that the provider rents instances that large. Serving many requests at once needs more memory for context.
Alternatives to the RTX 4070 SUPER
GPUs with similar compute and memory, and their lowest on-demand price right now.
- RTX 407012 GB · 0.0583 PF FP16$0.12/h$1.99/PFLOP·h
- RTX 3080 Ti12 GB · 0.0682 PF FP16$0.08/h$1.21/PFLOP·h
- RTX 4070 Ti12 GB · 0.0802 PF FP16$0.14/h$1.75/PFLOP·h
- RTX 308010 GB · 0.0595 PF FP16$0.12/h$2.02/PFLOP·h
- RTX 2080 Ti11 GB · 0.0538 PF FP16$0.08/h$1.42/PFLOP·h
- RTX 309024 GB · 0.071 PF FP16$0.09/h$1.27/PFLOP·h
RTX 4070 SUPER pricing questions
How much does it cost to rent an RTX 4070 SUPER?
On-demand RTX 4070 SUPER prices start at $0.09 per GPU-hour (Vast.ai). The median across 1 providers is $0.09, and the most expensive is $0.09. Running one GPU around the clock costs about $69 per month at the lowest price.
Is there cheaper spot or interruptible RTX 4070 SUPER capacity?
Yes. Spot (interruptible) RTX 4070 SUPER capacity starts at $0.12 per GPU-hour, -26% below the cheapest on-demand price. Spot instances can be preempted at any time, so use them for checkpointed training or batch jobs.
How much compute do you get per dollar with the RTX 4070 SUPER?
At $0.09 per hour and 0.071 PF of dense FP16/BF16, the RTX 4070 SUPER costs $1.34 per PFLOP-hour. That ranks 6 of 75 GPUs with on-demand prices on PetaPrice. Figures are datasheet peaks; real utilization is typically 30–60%.
What are the RTX 4070 SUPER's key specs?
12 GB GDDR6X with 504 GB/s of memory bandwidth, Ada Lovelace architecture (2024), PCIe form factor, 220 W. Dense peak: 0.071 PF FP16 / BF16, 0.142 PF FP8, 0.284 PF INT8, 0.0355 PF TF32, 0.0355 PF FP32, 0.0006 PF FP64.
How current are these prices?
Prices were collected on Nov 11, 2026, 07:02 UTC from provider APIs and public catalogs and refresh about every 2 hours. They are indicative list prices; always confirm on the provider's website before booking.