omnicoder-9b-mxfp4

tianrui6641/omnicoder-9b-mxfp4

omnicoder-9b-mxfp4 is a machine learning model from tianrui6641. Use it for the machine learning task on the model card, and read the license before you ship it in a product.

Downloads · 30 days

9% of all-time downloads

All-time downloads

Public

Parameters

5.7B

25.7 GB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors7.7 GB · 100%

Parameter types

How the weights are stored.

U83.7B · 64%

Model card

From the Hugging Face model README

    ---

    library_name: transformers

    base_model: Tesslate/OmniCoder-9B

    tags:

    - fp-quant

    - mxfp4

    - colab

    - uploaded-from-notebook

    ---



    # tianrui6641/omnicoder-9b-mxfp4



    This checkpoint was generated from `Tesslate/OmniCoder-9B` with `FPQuantConfig(forward_dtype="mxfp4")`.



    Notes:

    - This notebook uses generic FP-Quant MXFP4, not the GPT-OSS-only native MXFP4 loader path.

    - If `PSEUDOQUANTIZATION = True`, the checkpoint emulates MXFP4 quantization behavior and is serializable, but it does not provide Blackwell-only MXFP4 kernel speedups.

    - The quantization config is stored in `config.json`.

Type: qwen3_5_text
Created: Mar 30, 2026
Updated: Mar 30, 2026