LLM Opex

Models

Qwen2-72B: GPU memory and hosting cost

Open weights · 72.71 billion parameters · 131,072-token context · licence other

No cloud sells this model per token: to use it you run it yourself, on GPU instances you rent.

Run it yourself

It needs 181.8 GB of GPU memory at bf16, the precision it ships in.

CloudCheapest instance that fitsGPUsAn hourA month, around the clock
Core42mi300x.1x1 × MI300X 192 GB$3.50$2,555
DeepInfrab300.1x1 × B300 270 GB$5.89$4,300
Nebiusgpu-b300-sxm.1gpu-24vcpu-346gb1 × B300 288 GB$9.50$6,935
Alibaba Cloudecs.gn8v-2x.8xlarge2 × GPU H 96 GB$14.38$10,500
Fireworks AIb300.1x1 × B300 288 GB$15.00$10,950
Crusoea100-sxm.8x8 × A100 80 GB$18.40$13,432
Google Clouda2-ultragpu-4g4 × A100 80 GB$20.28$14,801
AWSp4d.24xlarge8 × A100 40 GB$21.96$16,029
AzureStandard_ND96asr_v48 × A100 40 GB$27.20$19,854
Together AIhgx-h100.8gpu8 × H100 80 GB$31.92$23,302
OCIBM.GPU.A100-v2.88 × A100 80 GB$32.00$23,360

More from Qwen

Similar size, open weights

List prices collected 21 November 2026, refreshed every three hours. No negotiated rates or taxes.