laion/glm46-code-feedback-maxeps-131k

Hugging Face
TEXT GENERATIONPricing:Input $0.468 / Output $1.82Concurrent Unit Cost:1Model Size:8BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Dec 16, 2025License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Warm

The laion/glm46-code-feedback-maxeps-131k is an 8 billion parameter language model fine-tuned from Qwen/Qwen3-8B. This model specializes in processing and generating code-related content, having been trained on the open-athena/glm46-code-feedback-maxeps-131k dataset. With a context length of 32768 tokens, it is optimized for tasks requiring extensive code understanding and feedback. Its primary strength lies in code-centric applications, leveraging its specialized fine-tuning for improved performance in programming contexts.

Loading preview...

Model Overview

laion/glm46-code-feedback-maxeps-131k is an 8 billion parameter language model built upon the robust Qwen/Qwen3-8B architecture. This model has undergone specialized fine-tuning using the open-athena/glm46-code-feedback-maxeps-131k dataset, indicating a strong focus on code-related tasks and feedback generation.

Key Training Details

The model was trained with a learning rate of 4e-05 over 7 epochs, utilizing a total train batch size of 16 across 8 GPUs. It employs the AdamW_Torch_Fused optimizer with a cosine learning rate scheduler and a warmup ratio of 0.1. This configuration suggests a methodical approach to optimizing performance for its intended domain.

Potential Use Cases

Given its fine-tuning on a code-feedback dataset, this model is likely well-suited for applications such as:

  • Code generation and completion: Assisting developers in writing code.
  • Code review and feedback: Providing suggestions or identifying issues in code.
  • Code understanding: Analyzing and interpreting programming constructs.
  • Educational tools: Aiding in learning programming by offering explanations or corrections.