Qwen3.5 21B GLM 4.7 Flash Heretic Uncensored Thinking needs about 15.0 GB of memory at Q4_K_M (a 12.9 GB file) with an 8K context window, so it fits on a 16 GB GPU. This repository has 11 quants, from 8.4 GB (Q2_K) to 22.1 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 6.4 GB.
Qwen3.5 21B GLM 4.7 Flash Heretic Uncensored Thinking – 21 billion parameter dense model
This Qwen 3.5 model contains 20.8 billion parameters and supports a 262 144 token context window. It is fully uncensored and includes vision, coding, reasoning, roleplay, and tools abilities. The model is 33 % faster than the 27 B base and 33 % smaller in tensor count.
Good at
- Large context reasoning
- Vision and image processing
- Tool‑based code generation
Limits
- No built‑in vision support in llama.cpp
- Requires at least 8 k context window
- License is Apache‑2.0
Best for
- Long‑form creative writing
- Complex multi‑step reasoning tasks
- Interactive coding assistants
Who it’s for. Developers and researchers needing a large, uncensored language model with extended context and vision support. The model is suitable for applications that require deep reasoning and tool integration.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q2_K | 8.4 GB | 10.5 GB | Too big | Get file → | |
| Q3_K_S | 9.5 GB | 11.6 GB | Too big | Get file → | |
| Q3_K_M | 10.4 GB | 12.5 GB | Too big | Get file → | |
| Q3_K_L | 11.2 GB | 13.3 GB | Too big | Get file → | |
| IQ4_XS | 11.8 GB | 13.9 GB | Too big | Get file → | |
| Q4_K_S | 12.1 GB | 14.2 GB | Too big | Get file → | |
| Q4_K_M | 12.9 GB | 15.0 GB | Too big | Get file → | |
| Q5_K_S | 14.5 GB | 16.6 GB | Too big | Get file → | |
| Q5_K_M | 14.9 GB | 17.0 GB | Too big | Get file → | |
| Q6_K | 17.1 GB | 19.2 GB | Too big | Get file → | |
| Q8_0 | 22.1 GB | 24.2 GB | Too big | Get file → |
For 8 GB VRAM use Q4_K_S (12.14 GB file, ~14.2 GB memory). For 12 GB VRAM use Q5_K_S (14.49 GB file, ~16.6 GB memory). For 16 GB VRAM use Q6_K (17.09 GB file, ~19.2 GB memory). Estimates are approximate.
What GPU do you need for Qwen3.5 21B GLM 4.7 Flash Heretic Uncensored Thinking?
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 Q3_K_S (9.5 GB file, about 11.6 GB in use).
- 16 GB VRAM: the best quant that fits is Q4_K_M (12.9 GB file, about 15.0 GB in use).
- 24 GB VRAM: the best quant that fits is Q6_K (17.1 GB file, about 19.2 GB in use).
- 32 GB VRAM: the best quant that fits is Q8_0 (22.1 GB file, about 24.2 GB in use).
Which Qwen3.5 21B GLM 4.7 Flash Heretic Uncensored Thinking quant should you choose?
Q4_K_M is the usual starting point for Qwen3.5 21B GLM 4.7 Flash Heretic Uncensored Thinking: it balances file size and quality well. Here it is a 12.9 GB file.
The Q8_0 file is 22.1 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 Q2_K 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
Frequently asked questions about Qwen3.5 21B GLM 4.7 Flash Heretic Uncensored Thinking
How much VRAM does Qwen3.5 21B GLM 4.7 Flash Heretic Uncensored Thinking need?
About 15.0 GB at Q4_K_M with an 8K context window. That is the 12.9 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 21B GLM 4.7 Flash Heretic Uncensored Thinking on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 10.5 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Qwen3.5 21B GLM 4.7 Flash Heretic Uncensored Thinking GGUF?
It depends on the quant: Q2_K is 8.4 GB, Q3_K_S is 9.5 GB, Q3_K_M is 10.4 GB, Q3_K_L is 11.2 GB, IQ4_XS is 11.8 GB, Q4_K_S is 12.1 GB and 5 more.
What context length does Qwen3.5 21B GLM 4.7 Flash Heretic Uncensored Thinking support?
The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 1.6 GB at 8K and 6.4 GB at 32K.
Where can I download Qwen3.5 21B GLM 4.7 Flash Heretic Uncensored Thinking GGUF?
The files are on Hugging Face in the mradermacher/Qwen3.5-21B-GLM-4.7-Flash-Heretic-Uncensored-Thinking-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 21B GLM 4.7 Flash Heretic Uncensored Thinking 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 Q4_K_S quant, which requires about 14.2 GB of VRAM at 8k context.
Does the model support image input?
Yes, vision ability is detected, but llama.cpp may need additional configuration.
What license applies to this model?
The model is released under the Apache‑2.0 license.