Qwen/Qwen3.5-27B. In this branch, the full-precision Qwen 3.5 conversion preserves the native mtp.* tensors in a sidecar mtp.safetensors file so they can be used for native speculative decoding after MLX conversion.

Downloads last month
168
Safetensors
Model size
27B params
Tensor type
BF16
·
F32
·
MLX
Hardware compatibility
Log In to add your hardware

Quantized

Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for trevon/Qwen3.5-27B-MLX-MTP

Base model

Qwen/Qwen3.5-27B
Finetuned
(233)
this model