OpenAI on NVIDIA Ada

Can I run gpt-oss 20B-A3.6B on an L40S?

Yes. gpt-oss 20B-A3.6B at Q4_K_M uses 13.1 GB of the 44.3 GB available on a L40S, and runs at about 166 tokens per second. You can push the context to 128K.

Runs comfortably

13.1 GB of 44.3 GB · 29%
044 GB
Weights 11.9 GB
KV cache 0.4 GB
Runtime overhead 0.8 GB

gpt-oss 20B-A3.6B at Q4_K_M leaves 31.3 GB spare on a L40S. There is room to raise the context length or move up a quantisation level.

Generation166tok/s
Prompt processing10558tok/s
Max context128Ktokens
KV per 1K tokens0GB

Every quantisation of gpt-oss 20B-A3.6B on a L40S

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

QuantisationbpwWeights TotalVerdictMax ctxtok/s
FP16 / BF16 16.00 38.9 GB 40.1 GB Fits, but tight 98K 60.6
INT8 / W8A8 8.50 20.4 GB 21.6 GB Runs comfortably 128K 107
FP8 (E4M3) 8.00 19.5 GB 20.6 GB Runs comfortably 128K 112
Q6_K 6.56 16.0 GB 17.1 GB Runs comfortably 128K 131
Q5_K_M 5.67 13.8 GB 15.0 GB Runs comfortably 128K 148
Q5_K_S 5.52 13.4 GB 14.6 GB Runs comfortably 128K 151
AWQ 4-bit 4.25 11.9 GB 13.1 GB Runs comfortably 128K 165
GPTQ 4-bit 4.25 11.9 GB 13.1 GB Runs comfortably 128K 165
MXFP4 4.25 11.9 GB 13.1 GB Runs comfortably 128K 165
Q4_K_M 4.85 11.9 GB 13.1 GB Runs comfortably 128K 166
Q4_K_S 4.58 11.3 GB 12.4 GB Runs comfortably 128K 173
Q4_0 4.55 11.2 GB 12.4 GB Runs comfortably 128K 173
IQ4_XS 4.25 10.5 GB 11.7 GB Runs comfortably 128K 182
Q3_K_M 3.91 9.7 GB 10.9 GB Runs comfortably 128K 192
IQ3_M 3.70 9.2 GB 10.4 GB Runs comfortably 128K 199
IQ3_XXS 3.06 7.8 GB 8.9 GB Runs comfortably 128K 225
Q2_K 2.63 6.8 GB 8.0 GB Runs comfortably 128K 246
IQ2_XXS 2.06 5.5 GB 6.7 GB Runs comfortably 128K 281
IQ1_M 1.75 4.8 GB 5.9 GB Runs comfortably 128K 305

Also worth checking

gpt-oss 20B-A3.6B on every GPU
Full hardware tableEverything that runs on a L40S
Full model tableWhy the KV cache is the part that bites
Guide