lllqaq/Qwen2.5-Coder-7B-fim-v2-filtered-0316

Hugging Face
TEXT GENERATIONConcurrency Cost:1Model Size:7.6BQuant:FP8Ctx Length:32kPublished:Mar 19, 2026License:otherArchitecture:Transformer Warm

Qwen2.5-Coder-7B-fim-v2-filtered-0316 is a 7.6 billion parameter language model, fine-tuned by lllqaq from the Qwen2.5-Coder-7B-Instruct base model. This iteration is specifically trained on the fim_midtrain_v2_filtered dataset, indicating an optimization for fill-in-the-middle code completion tasks. With a context length of 32768 tokens, it is designed for robust code generation and understanding within large codebases.

Loading preview...

Qwen2.5-Coder-7B-fim-v2-filtered-0316: A Code-Focused Language Model

This model is a specialized version of the Qwen2.5-Coder-7B-Instruct base model, fine-tuned by lllqaq. It features 7.6 billion parameters and supports a substantial 32768-token context length, making it suitable for handling extensive code snippets and projects.

Key Characteristics

  • Base Model: Derived from Qwen/Qwen2.5-Coder-7B-Instruct, indicating a foundation in instruction-following and code-related tasks.
  • Fine-tuning: Specifically trained on the fim_midtrain_v2_filtered dataset, suggesting an emphasis on fill-in-the-middle (FIM) capabilities, crucial for code completion and insertion.
  • Training Configuration: Utilized a learning rate of 1e-05, a total training batch size of 128, and a cosine learning rate scheduler with 0.1 warmup ratio over 1 epoch.

Intended Use Cases

Given its fine-tuning on a FIM-focused dataset, this model is particularly well-suited for:

  • Code Completion: Generating relevant code suggestions within existing code structures.
  • Code Generation: Assisting developers by completing partial code or generating new code segments.
  • Code Refactoring: Potentially aiding in modifying and improving code by suggesting insertions or changes.

This model is built using Transformers 4.57.1 and PyTorch 2.6.0+cu124, ensuring compatibility with modern deep learning frameworks.