GPU MODELS / TITAN XP

TITAN Xp cloud GPU pricing

3 listings across 2 providers. Cheapest published rate: $0.048/GPU/hr at Vast.ai. Sorted by per-GPU hourly price. Every row links to the provider's own published price.

WHAT IS THE TITAN XP

The Titan Xp was 2017's prosumer flagship - the card a generation of early deep-learning papers was trained on. It has no tensor cores and GDDR5X bandwidth; in modern catalogs it is a legacy floor-price entry for small experiments and old code.

MAKER
NVIDIA
ARCHITECTURE
Pascal
LAUNCHED
2017
MEMORY
12 GB GDDR5X
MEMORY BANDWIDTH
548 GB/s
TENSOR COMPUTE
12.1 TFLOPS FP32; no tensor cores
POWER DRAW
250 W
INTERCONNECT
PCIe 3.0 x16

WHAT IT IS GOOD AT

Legacy deep-learning code
Pascal-era training scripts still run.
Small experiments
12 GB of slow-ish memory at floor prices.
Modern inference
Not a fit - no tensor cores.

WHO SHOULD NOT RENT THE TITAN XP

Anything using tensor-core or BF16/FP8 paths.
Production serving.
ProviderConfigVRAMLocation$/hr$/GPU/hrSource
Vast.aiVast.aiTitan Xp × 212 GBSouth Korea, KR$0.096$0.048receipt
QuickPodQuickPodTITAN Xp × 112 GBNot published$0.050$0.050receipt
Vast.aiVast.aiTitan Xp × 112 GBSouth Korea, KR$0.056$0.056receipt

THE TITAN XP MARKET, BY THE NUMBERS

PROVIDERS
2
CONFIGURATIONS
3
MEDIAN $/GPU/HR
$0.05
SPREAD
1.2x

2 providers publish 3 TITAN Xp configurations on the panel today. Published per-GPU hourly rates run from $0.048 (Vast.ai) to $0.06 (Vast.ai), a 1x spread between the cheapest and the most expensive published rate for the same chip. The median listing sits at $0.05/GPU/hr. 3 rows were reported in stock by the provider at last observation; the rest are published catalog rates. All rates are on-demand - committed-use and reserved pricing is excluded.

TITAN XP PRICE TREND, DAILY PANEL SNAPSHOTS

Across every TITAN Xp row on the panel, the daily floor moved from $0.054 to $0.048/GPU/hr over 9 days of tracking (-12.2%). Snapshots run daily; the full history is public in the repo.

TITAN XP PROVIDER NOTES
Vast.ai · 2 configs · 1 region · 2 in stock
Sets the panel floor for this model.
from $0.048/hr
QuickPod · 1 config · 1 in stock
Within 15% of the panel floor.
from $0.050/hr
BUYING TITAN XP: WHAT THE PANEL SAYS

At the current floor, one TITAN Xp running around the clock costs about $35 a month (730 hours of on-demand arithmetic). Per-GPU floors by config size: 1x $0.05 · 2x $0.05. Compare the alternatives below before committing - the same budget often buys more than one chip class.

TITAN XP VS THE ALTERNATIVES

TITAN Xp vs P100
Same Pascal generation: the P100 adds HBM2 bandwidth and FP16 support, the Titan adds nothing but availability. For old code either works; neither suits modern stacks.
TITAN Xp vs RTX 3060
Same 12 GB, three generations newer, with tensor cores, usually at similar rates. The 3060 wins for any current use.

Compare live rates:

TITAN XP PRICING FAQ

What is the cheapest TITAN Xp cloud GPU right now?
The cheapest published TITAN Xp rate on the panel is $0.048 per GPU-hour at Vast.ai (Titan Xp x 2, South Korea, KR). Every price links to the provider's own published rate.
How much does a TITAN Xp cost per hour?
Across 2 providers tracked today, published per-GPU hourly rates for TITAN Xp run from $0.05 to $0.06, with a median of $0.05. Committed-use and reserved rates are excluded; everything shown is on-demand.
How many providers rent TITAN Xp GPUs?
2 providers publish 3 distinct TITAN Xp configurations on Compute Cafe. 3 of those rows were reported in stock by the provider's live endpoint at last observation.
Are TITAN Xp prices going up or down?
Over the 9 days Compute Cafe has tracked the panel, the TITAN Xp floor is falling: $0.054 to $0.048/GPU/hr (-12.2%). Daily snapshots are public in the repo, so the trend is auditable.
What does a TITAN Xp cost per month?
At the current floor of $0.048/GPU/hr, a TITAN Xp running 24/7 costs about $35 per month (730 hours of on-demand arithmetic on the cheapest published rate). Committed-use contracts can price lower; the panel tracks on-demand rates only.
Is the Titan Xp still usable for ML?
Only for Pascal-era code and small experiments. No tensor cores and GDDR5X put it far behind even budget modern cards.

See also alternatives to the TITAN Xp · cheapest GPU-hour per model · inference chips · training chips · by location