scottlowry
/

Qwen3.5-122B-A10B-oQ4e-mtp

4-bit precision

Model card Files Files and versions

Qwen3.5-122B-A10B-oQ4e-mtp

This model was quantized using oQ (oMLX v0.5.3) mixed-precision quantization.

Quantization details

Model type: qwen3_5_moe
Bits: 4
Group size: 64
Format: MLX safetensors

Downloads last month: 82

Safetensors

Model size

125B params

Tensor type

U32

·

BF16

·

MLX

Hardware compatibility

Log In to add your hardware

4-bit

Inference Providers NEW

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

Model tree for scottlowry/Qwen3.5-122B-A10B-oQ4e-mtp

Base model

Qwen/Qwen3.5-122B-A10B

Quantized

(153)

this model

Collection including scottlowry/Qwen3.5-122B-A10B-oQ4e-mtp

Qwen3.5-122B-A10B oQe MTP

6 items • Updated Sep 5