Qwen3.6 35B A3B REAM 192 heretic APEX needs about 14.5 GB of memory at Q3_K_L (a 13.4 GB file) with an 8K context window, so it fits on a 16 GB GPU. This repository has 5 quants, from 11.1 GB (IQ2_S) to 19.5 GB (Q5_K_M). Longer context needs more: at 32K tokens the KV cache alone adds about 2.7 GB.
Qwen3.6 35B A3B REAM 192 heretic APEX – a mixed‑precision MoE model
Qwen3.6 35B A3B REAM 192 heretic APEX is a 25‑billion‑parameter Qwen model with 3 billion active experts. It supports vision and is released under Apache‑2.0. The model is quantized with APEX for llama.cpp compatibility.
Good at
- Vision tasks
- Large context handling
- Mixture‑of‑experts inference
Limits
- Requires 8‑k context for memory estimates
- No official benchmark data
- Large VRAM needed for full precision
Best for
- Image‑guided text generation
- Long‑form document analysis
- Multimodal chat
Who it’s for. Developers needing a vision‑capable MoE model for llama.cpp. Suitable for research or production where mixed‑precision quantization is acceptable.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| IQ2_S | 11.1 GB | 12.3 GB | Too big | Get file → | |
| Q3_K_L | 13.4 GB | 14.5 GB | Too big | Get file → | |
| Q3_K_L | 13.4 GB | 14.5 GB | Too big | Get file → | |
| Q5_K_M | 17.6 GB | 18.8 GB | Too big | Get file → | |
| Q5_K_M | 19.5 GB | 20.7 GB | Too big | Get file → |
For 8 GB VRAM, use IQ2_S (11.08 GB file, ~12.3 GB memory). For 12 GB VRAM, Q3_K_L (13.37 GB file, ~14.5 GB memory). Higher‑quality options like Q5_K_M need 18–20 GB VRAM.
What GPU do you need for Qwen3.6 35B A3B REAM 192 heretic APEX?
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: the best quant that fits is Q3_K_L (13.4 GB file, about 14.5 GB in use).
- 24 GB VRAM: the best quant that fits is Q5_K_M (19.5 GB file, about 20.7 GB in use).
Which Qwen3.6 35B A3B REAM 192 heretic APEX quant should you choose?
Q3_K_L (13.4 GB) is the middle option in this repository. A reasonable choice when a 4-bit file will not fit, mostly on larger models. Quality loss is noticeable but often acceptable.
If memory is tight, the smallest file is IQ2_S at 11.1 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
Frequently asked questions about Qwen3.6 35B A3B REAM 192 heretic APEX
How much VRAM does Qwen3.6 35B A3B REAM 192 heretic APEX need?
About 14.5 GB at Q3_K_L with an 8K context window. That is the 13.4 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 35B A3B REAM 192 heretic APEX on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 12.3 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Qwen3.6 35B A3B REAM 192 heretic APEX GGUF?
It depends on the quant: IQ2_S is 11.1 GB, Q3_K_L is 13.4 GB, Q3_K_L is 13.4 GB, Q5_K_M is 17.6 GB, Q5_K_M is 19.5 GB.
What context length does Qwen3.6 35B A3B REAM 192 heretic APEX support?
The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 0.7 GB at 8K and 2.7 GB at 32K.
Where can I download Qwen3.6 35B A3B REAM 192 heretic APEX GGUF?
The files are on Hugging Face in the keithnull/Qwen3.6-35B-A3B-REAM-192-heretic-APEX-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 35B A3B REAM 192 heretic APEX 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 this model on 8 GB VRAM?
Use the IQ2_S quantization; it requires about 12.3 GB of VRAM at 8‑k context, which is close to 8 GB but may need swapping.
Does this model support vision?
Yes, the model has detected vision abilities and can process image inputs.
What license applies?
The model is released under the Apache‑2.0 license.