QriticalBeta
Open Model Catalogue
Alibaba Qwen
Alibaba Qwen
VisionText generation

Qwen3-VL 32B

Qwen/Qwen3-VL-32B-Instruct

Vision-language flagship on a single 80 GB GPU

Deploy Qwen3-VL 32B nowView on HuggingFace
Parameters
32B · vision
Min VRAM
≈ 78 GB
Hardware
1 × 80 GB GPU
OpenAI-compatible API with a per-workload key — drop-in for existing SDKs.
Dedicated GPUs — your model, your instance, no shared queues.
Billed by the second with auto-shutdown — never pay for idle time you didn't plan.