Qwen on NVIDIA Blackwell

Can I run Qwen2.5 72B on an RTX 5060 Ti 16GB?

Not at Q4_K_M — it needs 44.6 GB against 14.2 GB available. You would need 4 of these cards.

Won't fit

44.6 GB of 14.2 GB · 313%
046 GB
Weights 41.2 GB
KV cache 2.5 GB
Runtime overhead 0.9 GB
Over the limit 30.4 GB

Short by 30.4 GB. You can run it with 21 of 80 layers on the RTX 5060 Ti 16GB and the rest in system RAM, at roughly 1.28 tok/s — usable for batch work, painful for chat. A smaller quantisation or a shorter context is usually the better trade.

Generation1.28tok/s
Prompt processing136tok/s
Max context0tokens
KV per 1K tokens0GB

Every quantisation of Qwen2.5 72B on a RTX 5060 Ti 16GB

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

QuantisationbpwWeights TotalVerdictMax ctxtok/s
FP16 / BF16 16.00 135.4 GB 138.8 GB Won't fit — 0.33
INT8 / W8A8 8.50 71.4 GB 74.8 GB Won't fit — 0.67
Q8_0 (GGUF) 8.50 71.4 GB 74.8 GB Won't fit — 0.67
FP8 (E4M3) 8.00 67.7 GB 71.1 GB Won't fit — 0.70
Q6_K 6.56 55.5 GB 58.9 GB Won't fit — 0.89
Q5_K_M 5.67 48.0 GB 51.4 GB Won't fit — 1.06
Q5_K_S 5.52 46.7 GB 50.1 GB Won't fit — 1.09
Q4_K_M 4.85 41.2 GB 44.6 GB Won't fit — 1.28
AWQ 4-bit 4.25 39.4 GB 42.8 GB Won't fit — 1.36
GPTQ 4-bit 4.25 39.4 GB 42.8 GB Won't fit — 1.36
MXFP4 4.25 39.4 GB 42.8 GB Won't fit — 1.36
Q4_K_S 4.58 39.0 GB 42.4 GB Won't fit — 1.37
Q4_0 4.55 38.8 GB 42.2 GB Won't fit — 1.38
IQ4_XS 4.25 36.3 GB 39.7 GB Won't fit — 1.49
Q3_K_M 3.91 33.6 GB 36.9 GB Won't fit — 1.65
IQ3_M 3.70 31.8 GB 35.2 GB Won't fit — 1.79
IQ3_XXS 3.06 26.6 GB 30.0 GB Won't fit — 2.30
Q2_K 2.63 23.1 GB 26.5 GB Won't fit — 2.87
IQ2_XXS 2.06 18.4 GB 21.8 GB Won't fit — 4.31
IQ1_M 1.75 15.9 GB 19.3 GB Won't fit — 5.82

Also worth checking

Qwen2.5 72B on every GPU
Full hardware tableEverything that runs on a RTX 5060 Ti 16GB
Full model tableWhy the KV cache is the part that bites
Guide