AgPerry/Qwen2.5-Coder-7B-Instruct-num07

TEXT GENERATIONConcurrency Cost:1Model Size:7.6BQuant:FP8Ctx Length:32kPublished:Mar 16, 2026License:otherArchitecture:Transformer Cold

AgPerry/Qwen2.5-Coder-7B-Instruct-num07 is a 7.6 billion parameter instruction-tuned causal language model, fine-tuned from Qwen/Qwen2.5-Coder-7B-Instruct. This model specializes in code generation and understanding, having been further trained on specific fill-in-the-middle (FIM) datasets. It is optimized for programming tasks, leveraging its 32K context length for complex coding scenarios.

Loading preview...

AgPerry/Qwen2.5-Coder-7B-Instruct-num07 Overview

This model is a specialized 7.6 billion parameter instruction-tuned language model, derived from the Qwen2.5-Coder-7B-Instruct base model. Its primary focus is on enhancing code generation and comprehension capabilities through targeted fine-tuning.

Key Capabilities

  • Code Generation: Optimized for generating code, building upon the strong coding foundation of the Qwen2.5-Coder series.
  • Fill-in-the-Middle (FIM): Further fine-tuned on fim_midtrain_v1, fim_midtrain_v2, fim_midtrain_v3_pairs, and fim_midtrain_v3_triples datasets, suggesting improved performance in code completion and in-filling tasks.
  • Instruction Following: Designed to follow instructions effectively for programming-related queries.
  • Extended Context: Benefits from a 32,768 token context window, allowing for processing and generating longer code snippets and more complex programming logic.

Training Details

The model was trained with a learning rate of 1e-05, using an AdamW optimizer and a cosine learning rate scheduler with a 0.1 warmup ratio. The training involved 1 epoch with a total batch size of 128 across 8 GPUs, indicating a focused refinement process on the specialized FIM datasets.

Good For

  • Code Completion: Excels in scenarios requiring the model to complete partial code or fill in missing sections within existing codebases.
  • Code Generation Tasks: Suitable for generating new code based on natural language prompts or specific programming requirements.
  • Developer Tools: Can be integrated into IDEs or other developer tools to assist with coding, refactoring, and debugging support.

Finetunes
1 models