RTX PRO 6000 Blackwell
96GB GDDR7 per GPU
View GPU All GPUsL40S
Ada Lovelace architecture48GB GDDR6 per GPU
GPU Catalog / NVIDIA
L40S
GPU infrastructure.
A data-center GPU that spans generative AI, multimodal workloads, and high-end visual computing.
- Architecture
- Ada Lovelace
- GPU memory
- 48GB GDDR6 per GPU
- Published from
- $1.00 / hr
Published pricing
Available configurations.
Current Geodd hourly rates. GPU memory totals reflect the accelerator count in each published configuration.
| Configuration | Variant | Total GPU memory | Hourly rate | Action |
|---|---|---|---|---|
1x GPUgpu_1x_l40s | L40S | 48GB GDDR648GB GDDR6 each | $1.00per hour | Deploy |
2x GPUgpu_2x_l40s | L40S | 96GB GDDR648GB GDDR6 each | $2.00per hour | Deploy |
4x GPUgpu_4x_l40s | L40S | 192GB GDDR648GB GDDR6 each | $4.00per hour | Deploy |
8x GPUgpu_8x_l40s | L40S | 384GB GDDR648GB GDDR6 each | $8.00per hour | Deploy |
Hardware profile
Where L40S fits.
L40S combines Ada Lovelace architecture with 48GB of GDDR6 memory. NVIDIA positions it across LLM training and inference, multimodal generative AI, rendering, and real-time virtual production.
Choose L40S for a mixed AI and visualization workload that fits within 48GB per GPU and benefits from one platform for compute and graphics.
NVIDIA product information- 01
LLM training and inference
Run language-model development and serving workloads on a data-center GPU with 48GB of local memory.
- 02
Multimodal generative AI
Support pipelines that combine text, image, video, or other modalities within one accelerated environment.
- 03
Rendering and virtual production
Use Ada-generation rendering capability for interactive graphics and real-time production workflows.
Deployment sizing
Size before you provision.
Hardware fit depends on the model, precision, batch size, software stack, and total working set. Published use cases are not performance guarantees.
- 01 / Memory
Confirm the working set
Include weights, caches, activations, batches, and framework overhead.
- 02 / Software
Validate the stack
Check CUDA, driver, framework, precision, and container requirements.
- 03 / Capacity
Confirm provisioning
Published pricing does not guarantee immediate capacity in every configuration.