Qwen2.5 72B Instruct · 4× NVIDIA GeForce RTX 4090 · fp8

$/hr
$1.40
Weights / VRAM
73 / 96 GB
FP16 TFLOPS
660
HF repo
Qwen/Qwen2.5-72B-Instruct
Params
72.7B · qwen2
Context
131,072
License
qwen
Source
model: src · gpu: knowledge

← all offerings · deployments