✅Runs easilyfp16 · 4K ctx · batch 1
Run Llama 3.1 8B on L40S
At fp16 precision, Llama 3.1 8B (8.0 B parameters) needs roughly 17.7 GB of VRAM. L40S has 48 GB, leaving 30.3 GB of headroom — plenty of room for longer contexts and larger batch sizes. Expected throughput ≈ 43 tokens/sec on a single card.
37%
VRAM
Runs easily
Estimated VRAM (fp16)
17.7 GB
of 48 GB on L40S· 30.3 GB free
Tok/sec
43
TFTT
538ms
Power
168W
Rent on GPUniq marketplace
L40S
Live prices from verified providers · billed hourly, no commitment.
Memory breakdown (fp16)
Quantization comparison
Lower precision = less VRAM with a small quality trade-off. Quality order: FP16 > INT8 > INT4.
| Precision | VRAM | Utilisation | Fits on L40S? |
|---|---|---|---|
| FP16 (full precision) | 17.7 GB | 37% | ✅Runs easily |
| INT8 (8-bit) | 9.9 GB | 21% | ✅Runs easily |
| INT4 / Q4 (4-bit) | 6.0 GB | 12% | ✅Runs easily |
Frequently asked
- Can L40S run Llama 3.1 8B?
- Yes. Llama 3.1 8B needs ≈ 17.7 GB VRAM at fp16 and L40S provides 48 GB. Expected throughput is 43 tokens/sec per GPU.
- How much VRAM does Llama 3.1 8B use?
- About 17.7 GB at fp16, 6.0 GB at INT4 (for a 4K context, batch size 1). Longer contexts add to KV cache size; larger batches increase activations.
- What's the fastest way to run Llama 3.1 8B on L40S?
- Use a production inference engine like vLLM, SGLang, or TensorRT-LLM with paged attention — they cut KV cache 2–3× vs the naive estimate and batch multiple requests efficiently. For L40S, enable Flash Attention 2.
- Where can I rent a L40S?
- GPUniq aggregates live L40S offers from verified providers. You can deploy an instance in about a minute and pay hourly with no commitment.
Other GPUs for Llama 3.1 8B
Try it live
Chat with Llama 3.1 8B — no setup
Send a prompt and see how Llama 3.1 8B responds directly in our chat. No installation, no GPU required to test.
Want to tweak sequence length, batch size, or fine-tuning? Open the full calculator →