What GPU do I need to run Qwen/Qwen3.5-397B-A17B?
A 403B (MoE) language model for chat and instruction-following. 403.4B parameters, published in BF16. View on Hugging Face
Qwen3.5-397B-A17B is published by Qwen on Hugging Face, with 187,131 downloads and 1,556 likes to date. It's a Qwen3_5MoeForConditionalGeneration model built for image-text-to-text, published natively in BF16.
What Qwen3.5-397B-A17B is
Qwen3.5-397B-A17B is a 403B-parameter mixture-of-experts language model published by Alibaba (Qwen) on Hugging Face. It is released under Apache 2.0.
License note: fully permissive, no gating and no usage restrictions. Facts in this section are sourced from Qwen3.5-397B-A17B's Hugging Face model card, not benchmarked by Aquanode.
What it's used for
- Chat assistants
- Instruction following
- Synthetic data generation
VRAM required & cheapest live GPU fit
Required VRAM = weight size at each precision, plus a fixed overhead for KV-cache, activations, and fragmentation. Full formula and assumptions: methodology.
| Precision | Weight size | Required VRAM | Cheapest live fit | GPUs needed | Est. $/hr (full fit) |
|---|---|---|---|---|---|
| BF16 | 751.4 GB | 901.7 GB | No capable live offer found | – | – |
| FP8 (quantized) | 375.7 GB | 450.8 GB | RTX PRO 6000 | 5 | $6.88/hr |
| INT4 (quantized) | 187.8 GB | 225.4 GB | RTX A6000 | 5 | $1.81/hr |
A GPU is only matched to a row if its hardware supports that precision, and the primary recommendation is always a single-GPU fit when one exists.
INT4 requires a quantized checkpoint actually published for this model, check its Hugging Face page before relying on this row.
Qwen3.5-397B-A17B: common questions
Can Qwen3.5-397B-A17B run on a single GPU?
Not on a desktop card. At BF16 it needs 901.7 GB of VRAM, more than a single 32 GB desktop card holds. No card currently listed on the marketplace both supports BF16 and has enough VRAM for it, so how many it would take is not something this page can answer today.
How to run Qwen3.5-397B-A17B
Run Qwen3.5-397B-A17B with vLLM
Generic example, not from the model's own docs: adjust flags (quantization, context length, parallelism) for your setup.
vllm serve Qwen/Qwen3.5-397B-A17B --tensor-parallel-size 1Run Qwen3.5-397B-A17B with Ollama
Verified against Ollama's own library listing.
ollama run qwen3.5:397bRun Qwen3.5-397B-A17B with GGUF quantizations
Prebuilt GGUF weights published at unsloth/Qwen3.5-397B-A17B-GGUF. Run with llama.cpp's llama-server or load the repo directly in LM Studio.
llama-server -hf unsloth/Qwen3.5-397B-A17B-GGUFSource: https://huggingface.co/unsloth/Qwen3.5-397B-A17B-GGUF
Deploy Qwen3.5-397B-A17B on Aquanode
Aquanode has no one-click deploy template for Qwen3.5-397B-A17B; you install the inference engine yourself with the commands below. Aquanode sells GPU pods billed per second, not a hosted inference API.
- Launch a bare GPU pod sized to the requirement above (902 GB VRAM or more).
- Open a terminal on the pod, or save one of the commands above as a startup script so it runs automatically the first time the pod boots.
- Run the command and connect to the resulting endpoint.
Launch a GPU podWeight-to-VRAM math, the fit rules, and how live prices are normalized: full methodology.
More Qwen3.5 models
- Qwen3.5-397B-A17B-FP8 (403.4B, F8_E4M3)
- Qwen3.5-122B-A10B-FP8 (125.1B, F8_E4M3)
- Qwen3.5-122B-A10B (125.1B, BF16)
- Qwen3.5-35B-A3B-FP8 (36.0B, F8_E4M3)
- Qwen3.5-35B-A3B (36.0B, BF16)
Related reading: H100 pricing and specs, The best GPUs for AI, ranked, and Best GPU for LLM inference.
