yueqis/non_web-qwen-coder-32b-3epochs-30k-5e-5

Hugging Face
TEXT GENERATIONConcurrency Cost:2Model Size:32.8BQuant:FP8Ctx Length:32kPublished:Oct 24, 2025License:otherArchitecture:Transformer Warm

The yueqis/non_web-qwen-coder-32b-3epochs-30k-5e-5 model is a 32.8 billion parameter language model, fine-tuned from Qwen/Qwen2.5-Coder-32B-Instruct. It is specifically adapted for code generation and understanding tasks, leveraging its base model's capabilities. This model was fine-tuned on the 'non_web' dataset, suggesting a specialization for non-web-related code or data. Its primary strength lies in its enhanced performance for coding tasks within its specialized domain.

Loading preview...

Model Overview

The yueqis/non_web-qwen-coder-32b-3epochs-30k-5e-5 is a 32.8 billion parameter language model, fine-tuned from the robust Qwen/Qwen2.5-Coder-32B-Instruct base model. This iteration has undergone specific fine-tuning on the non_web dataset, indicating a specialized focus beyond general web-related content.

Key Characteristics

  • Base Model: Qwen2.5-Coder-32B-Instruct, known for its coding capabilities.
  • Parameter Count: 32.8 billion parameters, offering significant capacity for complex tasks.
  • Context Length: Supports a substantial context window of 32768 tokens.
  • Specialized Fine-tuning: Trained for 3 epochs with a learning rate of 5e-05 and a total batch size of 512, specifically on the non_web dataset.

Intended Use Cases

This model is particularly suited for applications requiring advanced code generation, comprehension, and analysis, especially within domains that align with its non_web fine-tuning data. Developers looking for a powerful coding assistant with a focus on non-web specific programming challenges may find this model beneficial.