machalek29/qwen3-0.6b-state-lifetime-tutor-n125

TEXT GENERATIONPricing:Input $0.04 / Cached $0.008 / Output $0.08Concurrent Unit Cost:1Model Size:0.8BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Aug 18, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

The machalek29/qwen3-0.6b-state-lifetime-tutor-n125 is a 0.8 billion parameter model based on Qwen3-0.6B, fine-tuned to act as a Python state-lifetime tutor. It is specifically designed to identify mutable-state lifetime bugs in short Python programs and ask a single, non-compound question about object creation, ownership, or shared references. This model excels at focused, diagnostic questioning without providing direct code corrections, operating with a context length of 32768 tokens.

Loading preview...

Python State-Lifetime Tutor (n=125)

This model, machalek29/qwen3-0.6b-state-lifetime-tutor-n125, is a specialized 0.8 billion parameter language model built upon the Qwen/Qwen3-0.6B base. Its unique purpose is to function as a Python state-lifetime tutor, identifying specific mutable-state lifetime bugs within short Python programs.

Key Capabilities

  • Bug Identification: Pinpoints relevant declarations, assignments, or mutations related to a single mutable-state lifetime bug.
  • Focused Questioning: Asks precisely one non-compound question concerning object creation, ownership, or shared references.
  • No Direct Corrections: Explicitly designed not to emit corrected code or state the correction, even when prompted.
  • System Prompt Adherence: Requires a specific system prompt and greedy decoding (do_sample=False) for intended behavior, as it was trained under these conditions.

Training and Evaluation

The model was trained using LoRA on the machalek29/state-lifetime-tutor-v1 dataset, specifically the first 125 ranked examples. Evaluation on 36 held-out scenarios (24 clean, 12 adversarial) showed 88% spec adherence on clean cases and 75% robustness on adversarial cases, with a 100% mechanical check pass rate.

Good For

  • Educational Tools: Ideal for integrating into educational platforms or IDEs where the goal is to guide users to discover solutions themselves rather than providing direct answers.
  • Diagnostic Assistance: Useful for developers seeking a focused, Socratic approach to understanding state-lifetime issues in their Python code.
  • Targeted Feedback: Provides precise, non-corrective feedback on specific types of Python bugs.

Finetunes
4 models