Qwen3.6 27B Omnimerge v4 needs about 19.2 GB of memory at Q4_K_M (a 16.5 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 27 quants, from 8.4 GB (IQ2_XXS) to 53.8 GB (F16). Longer context needs more: at 32K tokens the KV cache alone adds about 8.6 GB.
Qwen3.6 27B Omnimerge v4 – a 27B multimodal model with MLP‑passthrough surgery
Qwen3.6 27B Omnimerge v4 is a 27B‑parameter model that merges Qwen3.6 with three fine‑tunes. It supports coding, reasoning, and vision. The model is released under Apache‑2.0 and is available in 31 GGUF quantizations plus an MLX 4‑bit text‑only version.
Good at
- Coding tasks
- Complex reasoning
- Vision‑language input
Limits
- No vision support in MLX 4‑bit text‑only
- Large context requires high VRAM
- License is Apache‑2.0
Best for
- Code generation
- Multimodal document analysis
- Long‑form reasoning
Who it’s for. Developers and researchers needing a large multimodal model with proven reasoning and coding capabilities. The model is suitable for both interactive and batch workloads, especially when using the MTP companion for faster decoding.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| IQ2_XXS | 8.4 GB | 11.1 GB | Too big | Get file → | |
| IQ2_XS | 9.1 GB | 11.7 GB | Too big | Get file → | |
| IQ2_S | 9.4 GB | 12.0 GB | Too big | Get file → | |
| IQ2_M | 10.0 GB | 12.7 GB | Too big | Get file → | |
| Q2_K | 10.7 GB | 13.4 GB | Too big | Get file → | |
| IQ3_XXS | 11.2 GB | 13.8 GB | Too big | Get file → | |
| Q2_K_L | 12.0 GB | 14.6 GB | Too big | Get file → | |
| IQ3_XS | 12.0 GB | 14.6 GB | Too big | Get file → | |
| Q3_K_S | 12.1 GB | 14.7 GB | Too big | Get file → | |
| IQ3_M | 12.6 GB | 15.2 GB | Too big | Get file → | |
| Q3_K_M | 13.3 GB | 15.9 GB | Too big | Get file → | |
| Q3_K_L | 14.3 GB | 17.0 GB | Too big | Get file → | |
| Q3_K_XL | 14.4 GB | 17.1 GB | Too big | Get file → | |
| IQ4_XS | 15.1 GB | 17.7 GB | Too big | Get file → | |
| Q4_0 | 15.5 GB | 18.1 GB | Too big | Get file → | |
| Q4_K_S | 15.6 GB | 18.2 GB | Too big | Get file → | |
| IQ4_NL | 15.8 GB | 18.4 GB | Too big | Get file → | |
| Q4_K_M | 16.5 GB | 19.2 GB | Too big | Get file → | |
| Q4_1 | 17.1 GB | 19.7 GB | Too big | Get file → | |
| Q4_K_L | 17.5 GB | 20.1 GB | Too big | Get file → | |
| Q5_K_S | 18.7 GB | 21.3 GB | Too big | Get file → | |
| Q5_K_M | 19.2 GB | 21.9 GB | Too big | Get file → | |
| Q5_K_L | 20.0 GB | 22.7 GB | Too big | Get file → | |
| Q6_K | 22.1 GB | 24.7 GB | Too big | Get file → | |
| Q6_K_L | 22.7 GB | 25.3 GB | Too big | Get file → | |
| Q8_0 | 28.6 GB | 31.2 GB | Too big | Get file → | |
| F16 | 53.8 GB | 56.5 GB | Too big | Get file → |
For 8 GB VRAM, use IQ2_XXS (file size 8.43 GB, estimated 11.1 GB memory). For 12 GB VRAM, IQ2_M (10 GB, 12.7 GB memory) or Q2_K (10.71 GB, 13.4 GB memory). For 24 GB VRAM, Q6_K (22.08 GB, 24.7 GB memory) or Q6_K_L (22.7 GB, 25.3 GB memory).
What GPU do you need for Qwen3.6 27B Omnimerge v4?
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: the best quant that fits is IQ2_S (9.4 GB file, about 12.0 GB in use).
- 16 GB VRAM: the best quant that fits is Q3_K_M (13.3 GB file, about 15.9 GB in use).
- 24 GB VRAM: the best quant that fits is Q5_K_L (20.0 GB file, about 22.7 GB in use).
- 32 GB VRAM: the best quant that fits is Q8_0 (28.6 GB file, about 31.2 GB in use).
- 48 GB VRAM: the best quant that fits is Q8_0 (28.6 GB file, about 31.2 GB in use).
Which Qwen3.6 27B Omnimerge v4 quant should you choose?
Q4_K_M is the usual starting point for Qwen3.6 27B Omnimerge v4: it balances file size and quality well. Here it is a 16.5 GB file.
The Q8_0 file is 28.6 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 IQ2_XXS at 8.4 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 Qwen3.6 27B Omnimerge v4
How much VRAM does Qwen3.6 27B Omnimerge v4 need?
About 19.2 GB at Q4_K_M with an 8K context window. That is the 16.5 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 Qwen3.6 27B Omnimerge v4 on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 11.1 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Qwen3.6 27B Omnimerge v4 GGUF?
It depends on the quant: IQ2_XXS is 8.4 GB, IQ2_XS is 9.1 GB, IQ2_S is 9.4 GB, IQ2_M is 10.0 GB, Q2_K is 10.7 GB, IQ3_XXS is 11.2 GB and 21 more.
What context length does Qwen3.6 27B Omnimerge v4 support?
The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 2.1 GB at 8K and 8.6 GB at 32K.
Where can I download Qwen3.6 27B Omnimerge v4 GGUF?
The files are on Hugging Face in the ManniX-ITA/Qwen3.6-27B-Omnimerge-v4-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 Qwen3.6 27B Omnimerge v4 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 Qwen3.6 27B Omnimerge v4 on 8 GB VRAM?
Use the IQ2_XXS quant. It requires about 11.1 GB memory at 8 k context, which is close to 8 GB but may need swap or lower context.
Does the model support vision tasks?
Yes, the vision tower is preserved from the base model and works with the v4 GGUFs.
What is the MTP companion?
It is a weight‑identical GGUF with a draft‑MTP head for 2× faster decoding on a single 24 GB GPU.