Manual · GPU

L40S: Ada for AI

48GB, NVLink 3, and the FP8 gap. L40S is the Ada card that actually targets AI, not just graphics.

L40S is the Ada Lovelace card that bridges the gap between the consumer RTX line and the datacenter Hopper line. It's the "AI but not H100" option, and it's surprisingly capable.

What changed from L4

The numbers

Why it matters for inference

L40S is the "good enough for a 13B-30B model" card. 48GB fits a 70B model in INT4 (GPTQ/AWQ) on a single GPU, no tensor parallelism needed. Its FP8 support is real but not as complete as Hopper's Transformer Engine, so the FP8 win is smaller.

The interesting niche: L40S is often the cheapest way to get 48GB of GPU memory, making it a favorite for fine-tuning and for serving mid-size models where you need more memory than an L4 but don't want to pay H100 prices. Deploy it as independent replicas, not parallel-GPU training.

Sources

Back to the manual