01.AI on NVIDIA Ada

Can I run Yi-1.5 34B on an L40S?

Yes. Yi-1.5 34B at Q4_K_M uses 22.2 GB of the 44.3 GB available on a L40S, and runs at about 25.1 tokens per second. You can push the context to 32K.

Runs comfortably

22.2 GB of 44.3 GB · 50%
044 GB
Weights 19.5 GB
KV cache 1.9 GB
Runtime overhead 0.9 GB

Yi-1.5 34B at Q4_K_M leaves 22.1 GB spare on a L40S. There is room to raise the context length or move up a quantisation level.

Generation25.1tok/s
Prompt processing1105tok/s
Max context32Ktokens
KV per 1K tokens0GB

Every quantisation of Yi-1.5 34B on a L40S

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

QuantisationbpwWeights TotalVerdictMax ctxtok/s
FP16 / BF16 16.00 64.1 GB 66.8 GB Won't fit — 1.42
INT8 / W8A8 8.50 33.8 GB 36.6 GB Runs comfortably 32K 14.8
Q8_0 (GGUF) 8.50 33.8 GB 36.6 GB Runs comfortably 32K 14.8
FP8 (E4M3) 8.00 32.0 GB 34.8 GB Runs comfortably 32K 15.6
Q6_K 6.56 26.3 GB 29.0 GB Runs comfortably 32K 18.9
Q5_K_M 5.67 22.7 GB 25.5 GB Runs comfortably 32K 21.7
Q5_K_S 5.52 22.1 GB 24.9 GB Runs comfortably 32K 22.3
Q4_K_M 4.85 19.5 GB 22.2 GB Runs comfortably 32K 25.1
Q4_K_S 4.58 18.4 GB 21.2 GB Runs comfortably 32K 26.5
Q4_0 4.55 18.3 GB 21.1 GB Runs comfortably 32K 26.6
AWQ 4-bit 4.25 18.3 GB 21.0 GB Runs comfortably 32K 26.7
GPTQ 4-bit 4.25 18.3 GB 21.0 GB Runs comfortably 32K 26.7
MXFP4 4.25 18.3 GB 21.0 GB Runs comfortably 32K 26.7
IQ4_XS 4.25 17.2 GB 19.9 GB Runs comfortably 32K 28.3
Q3_K_M 3.91 15.8 GB 18.6 GB Runs comfortably 32K 30.6
IQ3_M 3.70 15.0 GB 17.8 GB Runs comfortably 32K 32.1
IQ3_XXS 3.06 12.5 GB 15.3 GB Runs comfortably 32K 38.0
Q2_K 2.63 10.8 GB 13.6 GB Runs comfortably 32K 43.3
IQ2_XXS 2.06 8.6 GB 11.4 GB Runs comfortably 32K 53.2
IQ1_M 1.75 7.4 GB 10.2 GB Runs comfortably 32K 60.7

Also worth checking

Yi-1.5 34B on every GPU
Full hardware tableEverything that runs on a L40S
Full model tableWhy the KV cache is the part that bites
Guide