llmrun

Qwen 3.6 Models — Hardware Requirements

14 Qwen 3.6 models from Alibaba and the community, from the smallest that runs in 2.7 GB of VRAM up to 36.0B parameters. Every row links to full quantization tables and GPU compatibility.

All Qwen 3.6 Models by Size

ModelParamsContext
Qwen3.6 27B MTPLX Optimized Speed4.7B262K
Qwen3.6 12B IQ Ultra Heretic Uncensored Thinking v2 Hightop12.1B262K
Qwen3.6 27B MTPLX Optimized26.9B262K
Qwen3.6 27B OBLITERATED26.9B262K
Qwen3.6 27B DFlash27B—
Qwen3.6 27B PRISM PRO DQ27B—
Qwen3.6 27B Uncensored Heretic v2 Native MTP Preserved27.4B262K
Qwen3.6 27B AEON Ultimate Uncensored BF1627.4B262K
Qwen3.6 27B Heretic2 Uncensored Finetune Thinking27.4B262K
Qwen3.6 27B27.8B262K
Huihui Qwen3.6 27B Abliterated27.8B262K
Qwen3.6 27B Uncensored HauhauCS Aggressive Safetensor Benchmark27.8B262K
Qwen3.6 28B REAP20 A3B28.2B262K
Qwen3.6 28B28.2B262K
Qwen3.6 35B A3B DFlash35B262K
Qwen3.6 35B A3b Crown Halo Mtp Dynamic35B—
Qwen3.6 35B A3B Uncensored Heretic Native MTP Preserved35.1B262K
Qwen3.6 35B A3B36.0B262K
Qwen3.6 35B A3B Claude 4.6 Opus Reasoning Distilled36.0B262K
Huihui Qwen3.6 35B A3B Claude 4.7 Opus Abliterated36.0B262K
Qwen3.6 35B A3B Claude 4.7 Opus Reasoning Distilled36.0B262K
Qwen3.6 40B Claude 4.6 Opus Deckard Heretic Uncensored Thinking39.5B262K

How Qwen 3.6 Compares — Benchmark Rating

Qwen3.6 27B is the highest-rated Qwen 3.6 model with an overall benchmark rating of 74.0/100 — #2 among 75 open models. The top proprietary model, GPT 5.5, scores 88.8. Click a model to see its full benchmark breakdown.

GPT 5.5 · proprietary88.8
Claude Opus 4.7 · proprietary87.6
Claude Fable 5 · proprietary86.6
GPT 5.4 · proprietary86.6
Claude Opus 4.8 · proprietary84.4
DeepSeek V4 Pro77.5
Qwen3.6 27B74.0
DeepSeek V4 Flash73.2
StableBeluga269.1
MiniMax M2.768.4
Kimi K2.668.2
Composite of normalized public benchmark scores (methodology) · ■ Qwen 3.6 · ■ other models

Frequently Asked Questions

How much VRAM do I need to run a Qwen 3.6 model?
The smallest Qwen 3.6 model, Qwen3.6 27B MTPLX Optimized Speed, runs from 2.7 GB of VRAM at an aggressive quantization. Larger family members need proportionally more — see the table above for every model.
Which Qwen 3.6 models can I run on a 16 GB GPU?
20 of 22 Qwen 3.6 models fit in 16 GB of VRAM at some quantization, including Qwen3.6 35B A3B, Qwen3.6 27B, Qwen3.6 27B Uncensored Heretic v2 Native MTP Preserved.
What is the most popular Qwen 3.6 model to run locally?
Qwen3.6 35B A3B is the most downloaded Qwen 3.6 model in local-friendly quantized formats. It runs from 10.3 GB of VRAM.
How do Qwen 3.6 models score on benchmarks?
Qwen3.6 27B leads the family with an overall benchmark rating of 74.0/100, ranking #2 among 75 open models, while the top proprietary model, GPT 5.5, scores 88.8. See the comparison chart above for the full standings.