Qwen1.5-110B-Chat needs about 70.4 GB of memory at Q4_K_M (a 67.2 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 14 quants, from 41.2 GB (Q2_K) to 118.2 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 10.7 GB.
Qwen1.5-110B-Chat is a 111B-parameter Qwen model in GGUF format, quantized by LiteLLMs. Its trained context length is 32K tokens. The repository has 14 quant files, from 41.2 GB (Q2_K) to 118.2 GB (Q8_0).
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q2_K | 41.2 GB (2 parts) | 44.4 GB | Too big | Get file → | |
| Q3_K_S | 48.5 GB (2 parts) | 51.6 GB | Too big | Get file → | |
| Q3_K_M | 53.7 GB (2 parts) | 56.9 GB | Too big | Get file → | |
| Q3_K_L | 58.1 GB (2 parts) | 61.3 GB | Too big | Get file → | |
| Q4_0 | 62.9 GB (2 parts) | 66.1 GB | Too big | Get file → | |
| Q4_K_S | 63.5 GB (2 parts) | 66.7 GB | Too big | Get file → | |
| Q4_K_M | 67.2 GB (3 parts) | 70.4 GB | Too big | Get file → | |
| Q4_1 | 69.8 GB (3 parts) | 72.9 GB | Too big | Get file → | |
| Q5_0 | 76.6 GB (3 parts) | 79.8 GB | Too big | Get file → | |
| Q5_K_S | 76.6 GB (3 parts) | 79.8 GB | Too big | Get file → | |
| Q5_K_M | 78.8 GB (3 parts) | 82.0 GB | Too big | Get file → | |
| Q5_1 | 83.5 GB (3 parts) | 86.7 GB | Too big | Get file → | |
| Q6_K | 91.2 GB (3 parts) | 94.4 GB | Too big | Get file → | |
| Q8_0 | 118.2 GB (4 parts) | 121.4 GB | Too big | Get file → |
What GPU do you need for Qwen1.5-110B-Chat?
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: the best quant that fits is Q2_K (41.2 GB file, about 44.4 GB in use).
Which Qwen1.5-110B-Chat quant should you choose?
Q4_K_M is the usual starting point for Qwen1.5-110B-Chat: it balances file size and quality well. Here it is a 67.2 GB file.
The Q8_0 file is 118.2 GB, about 1.8 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 Q2_K at 41.2 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 Qwen1.5-110B-Chat
How much VRAM does Qwen1.5-110B-Chat need?
About 70.4 GB at Q4_K_M with an 8K context window. That is the 67.2 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 Qwen1.5-110B-Chat on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 44.4 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Qwen1.5-110B-Chat GGUF?
It depends on the quant: Q2_K is 41.2 GB, Q3_K_S is 48.5 GB, Q3_K_M is 53.7 GB, Q3_K_L is 58.1 GB, Q4_0 is 62.9 GB, Q4_K_S is 63.5 GB and 8 more.
What context length does Qwen1.5-110B-Chat support?
The GGUF header lists a trained context length of 32,768 tokens (32K). The KV cache grows with the context you use: about 2.7 GB at 8K and 10.7 GB at 32K.
Where can I download Qwen1.5-110B-Chat GGUF?
The files are on Hugging Face in the LiteLLMs/Qwen1.5-110B-Chat-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 Qwen1.5-110B-Chat use?
Hugging Face lists the license as other. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.