https://recipes.vllm.ai/Qwen/Qwen3.5-397B-A17B
Qwen/Qwen3.5-397B-A17B | vLLM Recipes
Multimodal MoE model with gated delta networks architecture, 397B total / 17B active parameters, up to 262K context
qwenvllmrecipes
https://artificialanalysis.ai/models/comparisons/qwen3-5-397b-a17b-vs-step-3-5-flash-0202
Qwen3.5 397B A17B (Reasoning) vs Step 3.5 Flash: Model Comparison
Comparison between Qwen3.5 397B A17B (Reasoning) and Step 3.5 Flash across intelligence, price, speed, context window and more.
reasoning
https://endpoints.huggingface.co/new?accelerator=gpu&catalog_id=376&gguf_file=MXFP4_MOE%2FQwen3.5-397B-A17B-MXFP4_MOE-00001-of-00006.gguf&instance_id=aws-us-east-1-nvidia-a100-x4&no_suggested_compute=true®ion=us-east-1&repository=unsloth%2FQwen3.5-397B-A17B-GGUF&task=image-text-to-text&vendor=aws
Deploy unsloth/Qwen3.5-397B-A17B-GGUF | Inference Endpoints by Hugging Face
Deploy Qwen3.5-397B-A17B-GGUF for image-text-to-text inference in 1 click.