RTX 3090
See rental prices ↓Current median: $0.30/hr across 18 providers (observed), 0.00 (0.0%) vs yesterday. Typical range: $0.18–$0.42/hr (P10–P90). 30-day trend unlocks after 30 days of captures
TDP 350W · 71 TFLOPS · 24GB VRAM · 936 GB/s bandwidth
The RTX 3090 has a 350W TDP, 71 TFLOPS of FP16 compute, 24GB of GDDR6X VRAM, 936 GB/s memory bandwidth.
Previous-gen consumer GPU. Cheapest option for 24GB VRAM workloads. Limited BF16 support.
RTX 3090 currently rents for a median of $0.30/hr across 18 providers (observed on-demand prices). RTX 3090 has 24GB of memory. Compare its specs and listed cloud rental prices below. The workload must fit in VRAM with room for runtime memory; a listed price does not confirm stock.
RTX 3090 Overview
The RTX 3090 is an Ampere consumer GPU that remains relevant because it offers 24GB of GDDR6X at a comparatively low price. It provides 71 TFLOPS of FP16 performance and 936 GB/s of bandwidth, giving it considerably more memory headroom than many newer midrange consumer cards.
The 24GB pool can support smaller full-precision models and larger models after quantization, while 936 GB/s is respectable for its generation. Its shortcomings are equally important: it lacks ECC, does not provide NVLink memory pooling, and does not have native BF16 acceleration or Hopper-era FP8 support.
The card is a useful budget option for experimentation, small-model fine-tuning, and inference where 24GB matters more than newest-generation features. It is not intended for reliability-sensitive data-center deployments or for fast modern transformer training that relies on BF16 or FP8 paths.
Memory
Compute Performance
Hardware Specifications
Chip
Processors
Clocks
Memory
Power
Relative Performance
Relative to highest-spec GPU in database
Limitations
Live Cloud PricingOn-demand hourly rates
| Provider | On-Demand | Spot | Region | Availability |
|---|---|---|---|---|
BESTSwissGPU | $0.28/hr | — | CH | unknown |
| $0.45/hr | — | US/EU/Asia | unknown | |
| $0.26/hr | — | US/EU | unknown | |
| $0.23/hr | — | Asia-Pacific | unknown | |
| $0.21/hr | — | South Korea | unknown | |
| $0.29/hr | — | EU | unknown | |
| $0.20/hr | — | marketplace | unknown | |
| $0.38/hr | — | EU/US | unknown | |
| $0.31/hr | — | UA/PL/SE | unknown | |
| $0.22/hr | — | Global | unknown | |
| $0.29/hr | — | US | unknown | |
| $0.49/hr | — | US | unknown | |
| $0.30/hr | — | US | unknown | |
BESTNovita | $0.28/hr | — | US/EU/APAC | unknown |
| $0.22/hr | — | us-east-1 | unknown | |
| $0.32/hr | — | US | unknown | |
| $0.32/hr | — | Various | unknown | |
| $0.09/hr | — | US/EU | unknown | |
| $0.46/hr | — | US | unknown | |
| $0.23/hr | — | Minsk City | unknown | |
| $0.31/hr | — | US/EU/APAC | unknown | |
| $0.25/hr | — | US | unknown | |
| $0.30/hr | — | EU/US | unknown | |
| $0.40/hr | — | APAC | unknown | |
| $0.10/hr | — | Australia, AU | unknown | |
BESTHexabyte | $0.28/hr | — | EU | unknown |
| $0.35/hr | — | EU-North | unknown |
Salad: Lowest priority tier — preemptible
Job Cost Estimator
At median $0.30/hr across 18 providers (observed on-demand). Spot pricing typically 40–70% lower.
Estimates only. Actual cost depends on provider, region, availability, and utilization. Use the GPU cost calculator for custom inputs.
Compare RTX 3090 vs…
Use Case Guidance
LLM Model Size Guidance
Estimates only. Actual capacity depends on context length, KV cache, and framework overhead.
Related Guides
LLM APIs Running on This GPU Class
Providers that serve frontier LLM inference on Ampere-class hardware.
RTX 3090 vs Alternatives — Spec Comparison
| Spec | RTX 3090 this | RTX 5060 Ti | RTX 3080 Ti | RTX A5500 24GB |
|---|---|---|---|---|
| VRAM | 24GB GDDR6X▲ | 16GB GDDR7 | 12GB GDDR6X | 24GB GDDR6▲ |
| Memory Bandwidth | 936 GB/s▲ | 672 GB/s | 912 GB/s | 768 GB/s |
| FP16 TFLOPS | 71▲ | 70.4 | 68.2 | 68.2 |
| BF16 TFLOPS | 71▲ | 70.4 | 68.2 | 68.2 |
| FP8 TFLOPS | — | — | — | — |
| INT8 TOPS | 142 | 140.8 | 136 | 272.8▲ |
| TDP | 350W▲ | 180W | 350W▲ | 230W |
| Process Node | Samsung 8nm▲ | TSMC 4NP | Samsung 8nm▲ | Samsung 8nm▲ |
| Architecture | GA102 | GB206▲ | GA102 | GA102 |
| Release Year | 2020 | 2025▲ | 2021 | 2021 |
| Max model (FP16) | ~12B params▲ | ~8B params | ~6B params | ~12B params▲ |
| Max model (INT4) | ~48B params▲ | ~32B params | ~24B params | ~48B params▲ |
Compare RTX 3090
Related GPUs
RTX 3090 Frequently Asked Questions
What is the RTX 3090 TDP?
The RTX 3090 has a TDP (Thermal Design Power) of 350W.
How many TFLOPS does the RTX 3090 have?
The RTX 3090 delivers 71 TFLOPS of FP16 compute and 71 TFLOPS BF16.
How much VRAM does the RTX 3090 have?
The RTX 3090 has 24GB of GDDR6X VRAM.
What is the RTX 3090's memory bandwidth?
The RTX 3090 has a memory bandwidth of 936 GB/s.
RTX 3090 User Reviews
RTX 3090 User Reviews
Ready to rent?
Compare RTX 3090 prices across 130+ providers
Live on-demand & spot rates · monthly cost estimates · availability status