ModelUnvettedExternal

Qwen3.6 35B A3B OQ6e Mtp

This model was quantized using oQe (oMLX v0.5.0) imatrix-enhanced mixed-precision quantization.

Verification pending

Sourcehuggingface.co/jknlsn/Qwen3.6-35B-A3B-oQ6e-mtp

Kindimage-text-to-textBase modelQwen/Qwen3.6-35B-A3BVersionvb7ca41c3723da8c9df105c2eaabd5a20c200a061PublisherJknlsnDgrade

Model source

Kind: image-text-to-text
Base model: Qwen/Qwen3.6-35B-A3B
Version: vb7ca41c3723da8c9df105c2eaabd5a20c200a061
Publisher: Jknlsn
Source: Hugging Face

Source description

--- library_name: mlx pipeline_tag: image-text-to-text tags: - mlx - oq - oqe - imatrix - quantized - mtp base_model: Qwen/Qwen3.6-35B-A3B --- # Qwen3.6-35B-A3B-oQ6e-mtp This model was quantized using [oQe](https://github.com/jundot/omlx) (oMLX v0.5.0) imatrix-enhanced mixed-precision quantization. The model is converted from [`Qwen/Qwen3.6-35B-A3B`](https://huggingface.co/Qwen/Qwen3.6-35B-A3B) and includes multi-token prediction (MTP) support. Refer to the [original model card](https://huggingface.co/Qwen/Qwen3.6-35B-A3B) for more details on the model. ## Quantization details - **Model type**: qwen3_5_moe - **Bits**: 6 - **Group size**: 64 - **Enhanced (imatrix)**: yes - **Format**: MLX safetensors - **MTP**: included

Sources & provenance

1 active source

Hugging FaceRepresentative

huggingface.co/jknlsn/Qwen3.6-35B-A3B-oQ6e-mtp

highest_eligible_score

Verified Jul 11

Source evidence

3 excerpts

Readme excerptJul 11huggingface.co/jknlsn/Qwen3.6-35B-A3B-oQ6e-mtp

libraryname: mlx pipelinetag: image-text-to-text tags: mlx oq oqe imatrix quantized mtp basemodel: Qwen/Qwen3.6-35B-A3B

SummaryJul 11huggingface.co/jknlsn/Qwen3.6-35B-A3B-oQ6e-mtp