Qwen3.5-397B-A17B needs about 250.3 GB of memory at Q4_K_M (a 248.8 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 22 quants, from 121.1 GB (IQ1_M) to 806.2 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 4.1 GB.
Qwen3.5-397B-A17B: 397B parameter multimodal LLM with vision support
Qwen3.5-397B-A17B is a 397B parameter causal language model with vision encoder. It uses a qwen35moe architecture and supports a 262144 token context. The model is licensed under apache-2.0.
Good at
- Vision-language tasks
- Long context generation
- Multilingual support
Limits
- No explicit vision token limit listed
- Requires large VRAM for full precision
- No explicit tool‑calling limits
Best for
- Multimodal content creation
- Large‑scale inference on high‑end GPUs
- Research on vision‑language models
Who it’s for. Developers and researchers needing a large multimodal model for vision and language tasks. The model is suitable for environments that can accommodate its memory requirements and support the Qwen architecture.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| IQ1_M | 121.1 GB (4 parts) | 122.6 GB | Too big | Get file → | |
| IQ2_XXS | 126.4 GB (4 parts) | 127.9 GB | Too big | Get file → | |
| IQ2_M | 126.5 GB (4 parts) | 128.0 GB | Too big | Get file → | |
| Q2_K_XL | 134.7 GB (4 parts) | 136.2 GB | Too big | Get file → | |
| IQ3_XXS | 149.8 GB (4 parts) | 151.4 GB | Too big | Get file → | |
| IQ3_S | 164.0 GB (5 parts) | 165.5 GB | Too big | Get file → | |
| Q3_K_M | 182.6 GB (5 parts) | 184.1 GB | Too big | Get file → | |
| Q3_K_XL | 182.8 GB (5 parts) | 184.3 GB | Too big | Get file → | |
| IQ4_XS | 194.7 GB (5 parts) | 196.2 GB | Too big | Get file → | |
| IQ4_NL | 198.7 GB (5 parts) | 200.3 GB | Too big | Get file → | |
| Q4_K_S | 233.0 GB (6 parts) | 234.5 GB | Too big | Get file → | |
| MXFP4_MOE | 241.5 GB (6 parts) | 243.0 GB | Too big | Get file → | |
| Q4_K_M | 248.8 GB (7 parts) | 250.3 GB | Too big | Get file → | |
| Q4_K_XL | 249.6 GB (7 parts) | 251.1 GB | Too big | Get file → | |
| Q5_K_S | 281.8 GB (7 parts) | 283.3 GB | Too big | Get file → | |
| Q5_K_M | 299.7 GB (8 parts) | 301.2 GB | Too big | Get file → | |
| Q5_K_XL | 300.2 GB (8 parts) | 301.7 GB | Too big | Get file → | |
| Q6_K | 334.4 GB (8 parts) | 336.0 GB | Too big | Get file → | |
| Q6_K_XL | 366.2 GB (9 parts) | 367.7 GB | Too big | Get file → | |
| Q8_0 | 428.5 GB (10 parts) | 430.0 GB | Too big | Get file → | |
| Q8_K_XL | 436.7 GB (10 parts) | 438.3 GB | Too big | Get file → | |
| BF16 | 806.2 GB (18 parts) | 807.7 GB | Too big | Get file → |
For 8‑GB VRAM, use IQ1_M (121.07 GB file, ~122.6 GB memory at 8k context). For 16‑GB VRAM, IQ2_XXS (126.37 GB file, ~127.9 GB memory). For 32‑GB VRAM, Q4_K_M (248.79 GB file, ~250.3 GB memory). These estimates are approximate.
What GPU do you need for Qwen3.5-397B-A17B?
These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.
- 4 GB VRAM: no quant fits at an 8K context.
- 6 GB VRAM: no quant fits at an 8K context.
- 8 GB VRAM: no quant fits at an 8K context.
- 12 GB VRAM: no quant fits at an 8K context.
- 16 GB VRAM: no quant fits at an 8K context.
- 24 GB VRAM: no quant fits at an 8K context.
- 32 GB VRAM: no quant fits at an 8K context.
- 48 GB VRAM: no quant fits at an 8K context.
Which Qwen3.5-397B-A17B quant should you choose?
Q4_K_M is the usual starting point for Qwen3.5-397B-A17B: it balances file size and quality well. Here it is a 248.8 GB file.
The Q8_0 file is 428.5 GB, about 1.7 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.
If memory is tight, the smallest file is IQ1_M at 121.1 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about Qwen3.5-397B-A17B
How much VRAM does Qwen3.5-397B-A17B need?
About 250.3 GB at Q4_K_M with an 8K context window. That is the 248.8 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.
Can I run Qwen3.5-397B-A17B on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 122.6 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Qwen3.5-397B-A17B GGUF?
It depends on the quant: IQ1_M is 121.1 GB, IQ2_XXS is 126.4 GB, IQ2_M is 126.5 GB, Q2_K_XL is 134.7 GB, IQ3_XXS is 149.8 GB, IQ3_S is 164.0 GB and 16 more.
What context length does Qwen3.5-397B-A17B support?
The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 1.0 GB at 8K and 4.1 GB at 32K.
Where can I download Qwen3.5-397B-A17B GGUF?
The files are on Hugging Face in the unsloth/Qwen3.5-397B-A17B-MTP-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does Qwen3.5-397B-A17B use?
Hugging Face lists the license as apache-2.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.
Can I run Qwen3.5-397B-A17B on 8 GB VRAM?
The smallest quant, IQ1_M, requires about 122.6 GB of VRAM at 8k context, so 8 GB is insufficient.
Does this model support vision tasks?
Yes, the model has detected vision ability and includes a vision encoder.
What license applies to this model?
The model is released under the apache-2.0 license.