inferencerlabs/Qwen3.6-27B-MLX-9bit

Qwen3.6-27B

See Qwen3.6-27B in action: demonstration videos

For improved performance, please also download and enable a compatible MTP speculative decoder in inference settings.

Tested on a M3 Ultra 512GB RAM using Inferencer app

Text inference	~19.78 tokens/s @ 1000 tokens ~28.34 GiB (debug build)
Vision inference	~17.1 tokens/s ~28.36 GiB (debug build)
Vision inference (with MTP)	~29.04 tokens/s ~29.56 GiB (debug build)

Q9-bit quant typically achieves near lossless accuracy in our coding test

Quantization (bpw)	Perplexity	Token Accuracy	Missed Divergence
Q3.5	168.0	43.45%	72.57%
Q4.5	1.33593	91.65%	17.28%
Q5.5	1.23437	95.05%	17.28%
Q6.5	1.21875	96.65%	12.03%
Q8.5	1.21875	97.65%	9.92%
Q9	1.20312	97.80%	9.60%
Base	1.20312	100%	0.000%

• Perplexity: Measures the confidence for predicting base tokens (lower is better)
• Token Accuracy: The percentage of correctly generated base tokens
• Missed Divergence: Measures severity of misses; how much the token was missed by

Quantized with a modified version of MLX

For more details see demonstration videos or visit Qwen3.6-27B.

Disclaimer

We are not the creator, originator, or owner of any model listed. Each model is created and provided by third parties. Models may not always be accurate or contextually appropriate. You are responsible for verifying the information before making important decisions. We are not liable for any damages, losses, or issues arising from its use, including data loss or inaccuracies in AI-generated content.