Llama on NVIDIA Ada

Can I run Llama 3.2 3B on an L40S?

Yes. Llama 3.2 3B at Q4_K_M uses 3.5 GB of the 44.3 GB available on a L40S, and runs at about 219 tokens per second. You can push the context to 128K.

Runs comfortably

3.5 GB of 44.3 GB · 8%
044 GB
Weights 1.8 GB
KV cache 0.9 GB
Runtime overhead 0.8 GB

Llama 3.2 3B at Q4_K_M leaves 40.8 GB spare on a L40S. There is room to raise the context length or move up a quantisation level.

Generation219tok/s
Prompt processing11841tok/s
Max context128Ktokens
KV per 1K tokens0GB

Every quantisation of Llama 3.2 3B on a L40S

Highlighted row is the highest quality that still fits at 8K context.

QuantisationbpwWeights TotalVerdictMax ctxtok/s
FP16 / BF16 16.00 6.0 GB 7.7 GB Runs comfortably 128K 79.7
INT8 / W8A8 8.50 3.1 GB 4.8 GB Runs comfortably 128K 143
Q8_0 (GGUF) 8.50 3.1 GB 4.8 GB Runs comfortably 128K 143
FP8 (E4M3) 8.00 3.0 GB 4.7 GB Runs comfortably 128K 147
Q6_K 6.56 2.5 GB 4.1 GB Runs comfortably 128K 174
AWQ 4-bit 4.25 2.1 GB 3.8 GB Runs comfortably 128K 195
GPTQ 4-bit 4.25 2.1 GB 3.8 GB Runs comfortably 128K 195
MXFP4 4.25 2.1 GB 3.8 GB Runs comfortably 128K 195
Q5_K_M 5.67 2.1 GB 3.8 GB Runs comfortably 128K 196
Q5_K_S 5.52 2.1 GB 3.7 GB Runs comfortably 128K 200
Q4_K_M 4.85 1.8 GB 3.5 GB Runs comfortably 128K 219
Q4_K_S 4.58 1.8 GB 3.4 GB Runs comfortably 128K 227
Q4_0 4.55 1.7 GB 3.4 GB Runs comfortably 128K 228
IQ4_XS 4.25 1.6 GB 3.3 GB Runs comfortably 128K 239
Q3_K_M 3.91 1.5 GB 3.2 GB Runs comfortably 128K 252
IQ3_M 3.70 1.5 GB 3.1 GB Runs comfortably 128K 260
IQ3_XXS 3.06 1.3 GB 2.9 GB Runs comfortably 128K 291
Q2_K 2.63 1.1 GB 2.8 GB Runs comfortably 128K 316
IQ2_XXS 2.06 0.9 GB 2.6 GB Runs comfortably 128K 357
IQ1_M 1.75 0.8 GB 2.5 GB Runs comfortably 128K 384

Also worth checking

Llama 3.2 3B on every GPU
Full hardware tableEverything that runs on a L40S
Full model tableWhy the KV cache is the part that bites
Guide