Memory tier · 24 GB
Lowest-priced 24 GB GPU listings
5 GPU models carry 24 GB: RTX 3090, RTX 4090, L4, A10, and RTX A5000. The lowest-priced listing is $0.23 per GPU-hour for the RTX 3090 at Runcrate. Read 12 November 2026.
- Lowest listed
- $0.23
- Per GB-hour
- $0.010
- Providers
- 13
- One day
- $5.52
RTX 3090, Runcrate
lowest listed over memory
with a listing in the tier
one GPU at the lowest listing
Tier
GPU models with 24 GB
| GPU | Memory | Lowest-priced listing | $/GPU-hr | $/GB-hr | Providers | Source |
|---|---|---|---|---|---|---|
| RTX 3090consumer | 24 GB | Runcrateours, live | $0.23 | $0.010 | 4 | page |
| RTX 4090consumer | 24 GB | Novita AI | $0.33 | $0.014 | 8 | page |
| L4datacenter | 24 GB | RunPod | $0.59 | $0.025 | 5 | page |
| A10datacenter | 24 GB | OVHcloud | $1.00 | $0.042 | 4 | page |
| RTX A5000workstation | 24 GB | Runcrateours, live | $2.04 | $0.085 | 1 | page |
The tier is the card's stated memory, exactly 24 GB. Listings are provider-set on-demand prices in effect today, per GPU-hour. The rule for the lowest-priced listing is simply the lowest price among them.
Listings
Lowest-priced 24 GB listings, provider by provider
| # | GPU | Provider | Configuration | $/GPU-hr | Source |
|---|---|---|---|---|---|
| 1 | RTX 3090 | Runcrateours, live | 1× | $0.23 | page |
| 2 | RTX 3090 | RunC.AI | 1× · 1x RTX 3090 | $0.30 | page |
| 3 | RTX 3090 | SynpixCloud | 8× · 8x RTX 3090 | $0.31 | page |
| 4 | RTX 4090 | Novita AI | 1× · GPU instance, on-demand (sample configuration) | $0.33 | page |
| 5 | RTX 4090 | packet.ai | 1× · Dedicated (whole card) | $0.39 | page |
| 6 | RTX 4090 | Nova Cloud | 1× · 1x RTX 4090, 64 GB system RAM | $0.42 | page |
| 7 | RTX 4090 | RunC.AI | 1× · 1x RTX 4090 | $0.42 | page |
| 8 | RTX 4090 | CloudRift | 1× · On-demand (max 6 GPUs per instance) | $0.45 | page |
| 9 | RTX 4090 | SynpixCloud | 8× · 8x RTX 4090 | $0.47 | page |
| 10 | RTX 3090 | RunPod | 1× · Secure Cloud RTX 3090 | $0.50 | page |
| 11 | RTX 4090 | Runcrateours, live | 1× | $0.58 | page |
| 12 | L4 | RunPod | 1× · Secure Cloud L4 | $0.59 | page |
| 13 | L4 | Google Cloud | 1× · g2-standard-4 (4 vCPU, 16GiB RAM) | $0.71 | page |
| 14 | L4 | Amazon Web Services | 1× · g6.xlarge | $0.80 | page |
| 15 | RTX 4090 | RunPod | 1× · Secure Cloud RTX 4090 | $0.89 | page |
One line per provider and GPU, its lowest price. The fifteen lowest are shown of 22.
Or more
The lowest-priced ways to get at least 24 GB
| GPU | Memory | Lowest-priced listing | $/GPU-hr | $/GB-hr | Providers | Source |
|---|---|---|---|---|---|---|
| RTX 3090consumer | 24 GB | Runcrateours, live | $0.23 | $0.010 | 4 | page |
| V100datacenter | 32 GB | Verda | $0.24 | $0.008 | 3 | page |
| RTX 4090consumer | 24 GB | Novita AI | $0.33 | $0.014 | 8 | page |
| RTX A6000workstation | 48 GB | Thunder Compute | $0.35 | $0.007 | 7 | page |
| RTX 5090consumer | 32 GB | packet.ai | $0.36 | $0.011 | 8 | page |
| L40Sdatacenter | 48 GB | Runcrateours, live | $0.55 | $0.011 | 14 | page |
| A40datacenter | 48 GB | RunPod | $0.59 | $0.012 | 3 | page |
| L4datacenter | 24 GB | RunPod | $0.59 | $0.025 | 5 | page |
Every GPU model with 24 GB or more, ordered by lowest listed price, eight lowest. A bigger card can cost less than the card at exactly 24 GB; the memory is then free, but the card may be slower or older.
Arithmetic
What 24 GB holds
Weights alone, at 2 bytes per parameter (FP16), 24 GB holds about 12B parameters; at 1 byte (FP8), 24B; at half a byte (INT4), 48B. A 14 GB 7B model at FP16 fits in 24 GB, and a 26 GB 13B model at FP16 does not fit. These figures leave out the KV cache and activations, which take more memory as context and batch size grow, so treat them as a ceiling.
Questions
What is the lowest price for a 24 GB GPU?
$0.23 per GPU-hour for the RTX 3090 at Runcrate, read 2026-11-12.
Which GPUs have 24 GB?
RTX 3090 (4 providers), RTX 4090 (8 providers), L4 (5 providers), A10 (4 providers), and RTX A5000 (1 provider).
What does a 24 GB GPU cost per GB of memory?
At the lowest listing, $0.010 per GB-hour. The median of provider lows across the tier is $0.59 per GPU-hour.
Other memory tiers
Disclosure
Low-Cost GPU Finder is published by the Runcrate team, who also run Runcrate, an AI cloud. Runcrate is listed in the table, marked "ours, live". It is sorted by price like every other row, with no pinned position. Other providers' prices are read from their own public pricing pages. We have no paid placement.