Qwen1.5 55B Chat Cut i1 needs about 49.9 GB of memory at Q4_K_M (a 33.3 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 21 quants, from 12.5 GB (IQ1_S) to 45.0 GB (Q6_K). Longer context needs more: at 32K tokens the KV cache alone adds about 64.4 GB.
Qwen1.5 55B Chat Cut i1 – a 55B parameter chat model
Qwen1.5 55B Chat Cut i1 is a 55B parameter model derived from Qwen1.5-72B-Chat by removing 20 layers. It retains chat capabilities with some degradation in logic and coherence. The model supports a 32,768 token context window.
Good at
- Chatting with large context
- Generating long-form text
- Handling multi-turn dialogue
Limits
- Degraded logic compared to 72B
- No vision support
- License is other
Best for
- Developers needing a large chat model on limited GPU
- Researchers testing layer pruning effects
- Applications requiring 32k token context
Who it’s for. This model is suitable for developers and researchers who want to experiment with a large chat model that can run on GPUs with moderate VRAM. It is also useful for those studying the impact of layer removal on model performance.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| IQ1_S | 12.5 GB | 29.1 GB | Too big | Get file → | |
| IQ1_M | 13.5 GB | 30.1 GB | Too big | Get file → | |
| IQ2_XXS | 15.2 GB | 31.8 GB | Too big | Get file → | |
| IQ2_XS | 16.6 GB | 33.2 GB | Too big | Get file → | |
| IQ2_S | 17.9 GB | 34.5 GB | Too big | Get file → | |
| IQ2_M | 19.2 GB | 35.8 GB | Too big | Get file → | |
| Q2_K | 20.7 GB | 37.3 GB | Too big | Get file → | |
| IQ3_XXS | 21.1 GB | 37.7 GB | Too big | Get file → | |
| IQ3_XS | 22.8 GB | 39.4 GB | Too big | Get file → | |
| IQ3_S | 24.1 GB | 40.7 GB | Too big | Get file → | |
| Q3_K_S | 24.1 GB | 40.7 GB | Too big | Get file → | |
| IQ3_M | 25.3 GB | 41.9 GB | Too big | Get file → | |
| Q3_K_M | 26.8 GB | 43.4 GB | Too big | Get file → | |
| Q3_K_L | 29.3 GB | 45.9 GB | Too big | Get file → | |
| IQ4_XS | 29.5 GB | 46.1 GB | Too big | Get file → | |
| Q4_0 | 31.3 GB | 47.9 GB | Too big | Get file → | |
| Q4_K_S | 31.4 GB | 48.0 GB | Too big | Get file → | |
| Q4_K_M | 33.3 GB | 49.9 GB | Too big | Get file → | |
| Q5_K_S | 37.9 GB | 54.5 GB | Too big | Get file → | |
| Q5_K_M | 39.0 GB | 55.6 GB | Too big | Get file → | |
| Q6_K | 45.0 GB | 61.6 GB | Too big | Get file → |
For 8‑GB VRAM, use IQ1_S (file size 12.48 GB, ~29.1 GB memory at 8k context). For 12‑GB VRAM, IQ1_M (13.48 GB, ~30.1 GB). For 16‑GB VRAM, IQ2_XS (16.62 GB, ~33.2 GB). Memory estimates are approximate.
What GPU do you need for Qwen1.5 55B Chat Cut i1?
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: the best quant that fits is IQ2_XXS (15.2 GB file, about 31.8 GB in use).
- 48 GB VRAM: the best quant that fits is Q4_K_S (31.4 GB file, about 48.0 GB in use).
Which Qwen1.5 55B Chat Cut i1 quant should you choose?
Q4_K_M is the usual starting point for Qwen1.5 55B Chat Cut i1: it balances file size and quality well. Here it is a 33.3 GB file.
If memory is tight, the smallest file is IQ1_S at 12.5 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 55B Chat Cut i1
How much VRAM does Qwen1.5 55B Chat Cut i1 need?
About 49.9 GB at Q4_K_M with an 8K context window. That is the 33.3 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 55B Chat Cut i1 on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 29.1 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Qwen1.5 55B Chat Cut i1 GGUF?
It depends on the quant: IQ1_S is 12.5 GB, IQ1_M is 13.5 GB, IQ2_XXS is 15.2 GB, IQ2_XS is 16.6 GB, IQ2_S is 17.9 GB, IQ2_M is 19.2 GB and 15 more.
What context length does Qwen1.5 55B Chat Cut i1 support?
The GGUF header lists a trained context length of 32,768 tokens (32K). The KV cache grows with the context you use: about 16.1 GB at 8K and 64.4 GB at 32K.
Where can I download Qwen1.5 55B Chat Cut i1 GGUF?
The files are on Hugging Face in the mradermacher/Qwen1.5-55B-Chat-Cut-i1-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 55B Chat Cut i1 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.
Can I run this model on 8 GB VRAM?
Use the IQ1_S quant, which requires about 29.1 GB of VRAM at 8k context, so it will not fit on 8 GB VRAM.
Does this model support vision tasks?
No, the model has no vision capabilities.
What is the maximum context length?
The model supports up to 32,768 tokens.