⚙️ Chat template update (2026-09-14) — The tool-calling chat template has been replaced with the community-fixed froggeric v22.5 template, which fixes several tool-calling bugs (serialized JSON arguments, premature stops mid tool-call, reasoning_effort=high).

Qwen3.8-27B-oQ3.5-mtp

This model was quantized using oQ (oMLX v0.6.0.dev1) mixed-precision quantization.

Quantization details

Model type: qwen3_5
Bits: 3
Group size: 64
Format: MLX safetensors

Downloads last month: 102

Safetensors

Model size

28B params

Tensor type

U32

BF16

MLX

Hardware compatibility

3-bit

Inference Providers NEW

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

Model tree for d9beuD/Qwen3.8-27B-oQ3.5-mtp

Base model

Qwen/Qwen3.8-27B

Quantized

(1,767)

this model

Collection including d9beuD/Qwen3.8-27B-oQ3.5-mtp

Qwen3.8-27B oQ

Collection

9 items • Updated Aug 15