Instructions to use openjev/OpenJev-Flash-9B-MLX-4bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use openjev/OpenJev-Flash-9B-MLX-4bit with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] hf download openjev/OpenJev-Flash-9B-MLX-4bit --local-dir OpenJev-Flash-9B-MLX-4bit
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Atomic Chat
Download NOTICE from openjev/OpenJev-Flash-9B-MLX-4bit: direct link, hf CLI and curl.
- Browser
- Download file 860 Bytes
-
https://huggingface.co/openjev/OpenJev-Flash-9B-MLX-4bit/resolve/main/NOTICE
- Command line
-
hf download hf://openjev/OpenJev-Flash-9B-MLX-4bit/NOTICE
-
curl -L -o NOTICE https://huggingface.co/openjev/OpenJev-Flash-9B-MLX-4bit/resolve/main/NOTICE
860 Bytes
| OpenJev Flash 9B — release v1 | |
| OpenJev Flash 9B weights: Copyright the OpenJev project, released under CC BY-NC 4.0 (see LICENSE). | |
| These weights are a derivative of Qwen/Qwen3.5-9B, revision c202236235762e1c871ad0ccb60c8ee5ba337b9a, licensed under the Apache License, Version 2.0, and were produced by fine-tuning that base and merging the result into its weights. The MLX builds at openjev/OpenJev-Flash-9B-MLX (8-bit) and openjev/OpenJev-Flash-9B-MLX-4bit (4-bit) are conversions of these weights. | |
| Base model licence: https://huggingface.co/Qwen/Qwen3.5-9B/blob/c202236235762e1c871ad0ccb60c8ee5ba337b9a/LICENSE | |
| The base licence text is included as LICENSE-APACHE-2.0. | |
| The files in helper/ and serve/ are released under the Apache License, Version 2.0. The frozen helper is unchanged from OpenJev v1. | |
| For a commercial licence, email support@loopai.com. | |