Run HauhauCS/Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive locally

License: apache-2.0 ⬇ 3,089,944 ❤ 2328
Parameters34.66B
Context262,144

HauhauCS/Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive is a very large language model with 34.66 billion parameters, built on the qwen35moe architecture. It is released under the apache-2.0 license and has been downloaded 3,089,944 times.

To run HauhauCS/Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive locally at a 4,096-token context, its quantized versions need between 2.11 GB (F16, lowest quality) and 41.88 GB (Q8_K_P, highest quality) of memory, weights plus KV cache and a system margin included.

For most users the best balance is F16, needing about 2.11 GB. That means HauhauCS/Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive 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 0.21 Very low 0.84 GB 0.47 GB 2.11 GB 477.6 t/s Fits in VRAM
IQ2_M 2.69 Low 10.86 GB 0.47 GB 12.13 GB 4.6 t/s Offload
Q2_K_P 3.46 Fair 13.95 GB 0.47 GB 15.23 GB 3.6 t/s Offload
IQ3_M 3.56 Fair 14.38 GB 0.47 GB 15.65 GB 3.5 t/s Offload
IQ4_XS 4.32 Good 17.44 GB 0.47 GB 18.72 GB 2.9 t/s Offload
Q3_K_P 4.39 Good 17.72 GB 0.47 GB 18.99 GB 2.8 t/s Offload
IQ4_NL 4.57 Good 18.42 GB 0.47 GB 19.7 GB 2.7 t/s Offload
Q4_K_M 4.89 Good 19.71 GB 0.47 GB 20.99 GB 2.5 t/s Offload
Q4_K_P 5.41 Very good 21.82 GB 0.47 GB 23.09 GB 2.3 t/s Offload
Q5_K_P 6.47 Very good 26.1 GB 0.47 GB 27.38 GB Insufficient
Q6_K_P 7.07 Excellent 28.54 GB 0.47 GB 29.82 GB Insufficient
Q8_K_P 10.06 Excellent 40.61 GB 0.47 GB 41.88 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 HauhauCS/Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive?

You need about 2.11 GB of VRAM to run HauhauCS/Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive entirely on the GPU using the F16 quantization (at a 4,096-token context). Smaller quantizations lower the requirement at the cost of quality.

Can I run HauhauCS/Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive on an 8 GB GPU?

Yes. With 8 GB of VRAM you can run HauhauCS/Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive fully on the GPU using F16 (about 2.11 GB).

Can I run HauhauCS/Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive on a 16 GB GPU?

Yes. With 16 GB of VRAM you can run HauhauCS/Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive fully on the GPU using IQ3_M (about 15.65 GB).

Can I run HauhauCS/Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive on a 24 GB GPU?

Yes. With 24 GB of VRAM you can run HauhauCS/Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive fully on the GPU using Q4_K_P (about 23.09 GB).

What is the best quantization for HauhauCS/Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive?

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.