Run DavidAU/Qwen3.6-27B-Heretic-Uncensored-FINETUNE-NEO-CODE-Di-IMatrix-MAX-GGUF locally

License: apache-2.0 ⬇ 354,747 ❤ 356
Parameters26.9B
Context262,144

DavidAU/Qwen3.6-27B-Heretic-Uncensored-FINETUNE-NEO-CODE-Di-IMatrix-MAX-GGUF is a large code-focused language model with 26.9 billion parameters, built on the qwen35 architecture. It is released under the apache-2.0 license and has been downloaded 354,747 times.

To run DavidAU/Qwen3.6-27B-Heretic-Uncensored-FINETUNE-NEO-CODE-Di-IMatrix-MAX-GGUF locally at a 4,096-token context, its quantized versions need between 2.14 GB (F16, lowest quality) and 29.09 GB (Q8_0, highest quality) of memory, weights plus KV cache and a system margin included.

For most users the best balance is F32, needing about 2.99 GB. That means DavidAU/Qwen3.6-27B-Heretic-Uncensored-FINETUNE-NEO-CODE-Di-IMatrix-MAX-GGUF fits entirely in the VRAM of a 6 GB GPU or larger, running fully on the GPU.

→ Guide: How much VRAM do you need?

All quantizations

Quant.Bits QualityWeights KVTotal Speed~Verdict
F16 0.28 Very low 0.86 GB 0.47 GB 2.14 GB 463.0 t/s Fits in VRAM
BF16 0.28 Very low 0.87 GB 0.47 GB 2.14 GB 461.3 t/s Fits in VRAM
F32 0.55 Very low 1.72 GB 0.47 GB 2.99 GB 233.0 t/s Fits in VRAM
IQ2_M 3.12 Low 9.77 GB 0.47 GB 11.04 GB 5.1 t/s Offload
IQ3_M 3.83 Fair 12.01 GB 0.47 GB 13.28 GB 4.2 t/s Offload
IQ4_XS 4.58 Good 14.34 GB 0.47 GB 15.61 GB 3.5 t/s Offload
Q4_K_S 4.73 Good 14.81 GB 0.47 GB 16.08 GB 3.4 t/s Offload
IQ4_NL 4.79 Good 15.01 GB 0.47 GB 16.28 GB 3.3 t/s Offload
Q4_K_M 5.02 Very good 15.7 GB 0.47 GB 16.98 GB 3.2 t/s Offload
Q5_K_S 5.65 Very good 17.69 GB 0.47 GB 18.96 GB 2.8 t/s Offload
Q5_K_M 5.81 Very good 18.2 GB 0.47 GB 19.47 GB 2.7 t/s Offload
Q6_K 6.66 Excellent 20.86 GB 0.47 GB 22.13 GB 2.4 t/s Offload
Q8_0 8.88 Excellent 27.82 GB 0.47 GB 29.09 GB Insufficient

KV cache computed from the model's exact architecture. Speed is a rough estimate bounded by memory bandwidth.

Frequently asked questions

How much VRAM do you need to run DavidAU/Qwen3.6-27B-Heretic-Uncensored-FINETUNE-NEO-CODE-Di-IMatrix-MAX-GGUF?

You need about 2.99 GB of VRAM to run DavidAU/Qwen3.6-27B-Heretic-Uncensored-FINETUNE-NEO-CODE-Di-IMatrix-MAX-GGUF entirely on the GPU using the F32 quantization (at a 4,096-token context). Smaller quantizations lower the requirement at the cost of quality.

Can I run DavidAU/Qwen3.6-27B-Heretic-Uncensored-FINETUNE-NEO-CODE-Di-IMatrix-MAX-GGUF on an 8 GB GPU?

Yes. With 8 GB of VRAM you can run DavidAU/Qwen3.6-27B-Heretic-Uncensored-FINETUNE-NEO-CODE-Di-IMatrix-MAX-GGUF fully on the GPU using F32 (about 2.99 GB).

Can I run DavidAU/Qwen3.6-27B-Heretic-Uncensored-FINETUNE-NEO-CODE-Di-IMatrix-MAX-GGUF on a 16 GB GPU?

Yes. With 16 GB of VRAM you can run DavidAU/Qwen3.6-27B-Heretic-Uncensored-FINETUNE-NEO-CODE-Di-IMatrix-MAX-GGUF fully on the GPU using IQ4_XS (about 15.61 GB).

Can I run DavidAU/Qwen3.6-27B-Heretic-Uncensored-FINETUNE-NEO-CODE-Di-IMatrix-MAX-GGUF on a 24 GB GPU?

Yes. With 24 GB of VRAM you can run DavidAU/Qwen3.6-27B-Heretic-Uncensored-FINETUNE-NEO-CODE-Di-IMatrix-MAX-GGUF fully on the GPU using Q6_K (about 22.13 GB).

What is the best quantization for DavidAU/Qwen3.6-27B-Heretic-Uncensored-FINETUNE-NEO-CODE-Di-IMatrix-MAX-GGUF?

If memory allows, higher bits-per-weight means better quality. A common sweet spot is a Q4_K_M or Q5_K_M quantization, which keeps most of the quality while roughly halving the memory versus 8-bit. Pick the highest quantization that still fits in your VRAM.