Qwen on NVIDIA Blackwell

Can I run Qwen2.5-Coder 32B on an RTX 5070?

Not at Q4_K_M — it needs 21.5 GB against 10.5 GB available. You would need 3 of these cards.

Won't fit

21.5 GB of 10.5 GB · 205%
022 GB
Weights 18.6 GB
KV cache 2.0 GB
Runtime overhead 0.8 GB
Over the limit 11.0 GB

Short by 11.0 GB. You can run it with 26 of 64 layers on the RTX 5070 and the rest in system RAM, at roughly 3.40 tok/s — usable for batch work, painful for chat. A smaller quantisation or a shorter context is usually the better trade.

Generation3.40tok/s
Prompt processing397tok/s
Max context0tokens
KV per 1K tokens0GB

Every quantisation of Qwen2.5-Coder 32B on a RTX 5070

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

QuantisationbpwWeights TotalVerdictMax ctxtok/s
FP16 / BF16 16.00 61.0 GB 63.9 GB Won't fit — 0.77
INT8 / W8A8 8.50 32.1 GB 34.9 GB Won't fit — 1.62
Q8_0 (GGUF) 8.50 32.1 GB 34.9 GB Won't fit — 1.62
FP8 (E4M3) 8.00 30.5 GB 33.3 GB Won't fit — 1.73
Q6_K 6.56 25.0 GB 27.9 GB Won't fit — 2.22
Q5_K_M 5.67 21.6 GB 24.5 GB Won't fit — 2.70
Q5_K_S 5.52 21.1 GB 23.9 GB Won't fit — 2.83
Q4_K_M 4.85 18.6 GB 21.5 GB Won't fit — 3.40
AWQ 4-bit 4.25 18.3 GB 21.2 GB Won't fit — 3.45
GPTQ 4-bit 4.25 18.3 GB 21.2 GB Won't fit — 3.45
MXFP4 4.25 18.3 GB 21.2 GB Won't fit — 3.45
Q4_K_S 4.58 17.6 GB 20.5 GB Won't fit — 3.66
Q4_0 4.55 17.5 GB 20.4 GB Won't fit — 3.68
IQ4_XS 4.25 16.4 GB 19.3 GB Won't fit — 4.10
Q3_K_M 3.91 15.2 GB 18.0 GB Won't fit — 4.75
IQ3_M 3.70 14.4 GB 17.3 GB Won't fit — 5.11
IQ3_XXS 3.06 12.1 GB 15.0 GB Won't fit — 7.39
Q2_K 2.63 10.6 GB 13.4 GB Won't fit — 10.4
IQ2_XXS 2.06 8.5 GB 11.3 GB Won't fit 5K 22.7
IQ1_M 1.75 7.4 GB 10.2 GB Fits, but tight 9K 52.1

Also worth checking

Qwen2.5-Coder 32B on every GPU
Full hardware tableEverything that runs on a RTX 5070
Full model tableWhy the KV cache is the part that bites
Guide