Instructions to use RESMP-DEV/LFM2.5-Encoder-230M-Code-MXFP4-GPTQ with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use RESMP-DEV/LFM2.5-Encoder-230M-Code-MXFP4-GPTQ with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] hf download RESMP-DEV/LFM2.5-Encoder-230M-Code-MXFP4-GPTQ --local-dir LFM2.5-Encoder-230M-Code-MXFP4-GPTQ
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Atomic Chat
|
Download MODIFICATIONS.md from RESMP-DEV/LFM2.5-Encoder-230M-Code-MXFP4-GPTQ: direct link, hf CLI and curl.
- Browser
- Download file 569 Bytes
-
https://huggingface.co/RESMP-DEV/LFM2.5-Encoder-230M-Code-MXFP4-GPTQ/resolve/main/MODIFICATIONS.md
- Command line
-
hf download hf://RESMP-DEV/LFM2.5-Encoder-230M-Code-MXFP4-GPTQ/MODIFICATIONS.md
-
curl -L -o MODIFICATIONS.md https://huggingface.co/RESMP-DEV/LFM2.5-Encoder-230M-Code-MXFP4-GPTQ/resolve/main/MODIFICATIONS.md
569 Bytes
| # Modifications by RESMP.DEV | |
| This is a derivative of the identified LiquidAI LFM2.5 Encoder checkpoint, not an | |
| official LiquidAI release. RESMP.DEV removed the masked-language-model head and | |
| contrastively fine-tuned the complete encoder body for code retrieval using the procedure | |
| and corpus hashes in `training_report.json`. The resulting weights are stored in BF16. | |
| RESMP.DEV activation-calibrated the BF16 weights with block-GPTQ and packed eligible linear layers as native group-32 MLX MXFP4. Exact settings and hashes are recorded in `quantization_report.json`. | |