A small finetune over https://huggingface.co/datasets/Etherll/code-fim-v2 dataset on top of Qwen/Qwen2.5-Coder-1.5B to generate code FIM ( Fill-in-the-Middle ) You can use this with Continue.

Dont forget to use this format :

<|file_name|>{{{filename}}}<|fim_prefix|>{{{prefix}}}<|fim_suffix|>{{{suffix}}}<|fim_middle|>

Thanks

simplepod.ai for the compute

Downloads last month
185
Safetensors
Model size
1.54B params
Tensor type
BF16
·
Inference Providers NEW
This model is not currently available via any of the supported Inference Providers.

Model tree for Etherll/Qwen2.5-CodeFIM-1.5B-v2

Quantizations
2 models

Dataset used to train Etherll/Qwen2.5-CodeFIM-1.5B-v2