Keven16/Qwen3-4B-Non-Thinking-RL-Code-Step1200

TEXT GENERATIONPricing:Input $0.4 / Cached $0.08 / Output $0.8Concurrent Unit Cost:1Model Size:4BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Mar 16, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

Keven16/Qwen3-4B-Non-Thinking-RL-Code-Step1200 is a 4 billion parameter language model based on the Qwen3 architecture, fine-tuned for code generation tasks. This model is specifically optimized for producing code outputs, leveraging its training steps to enhance programming-related capabilities. With a context length of 32768 tokens, it is designed for developers requiring a focused and efficient code generation assistant.

Loading preview...

Model Overview

Keven16/Qwen3-4B-Non-Thinking-RL-Code-Step1200 is a 4 billion parameter model built upon the Qwen3 architecture. This model has undergone specific fine-tuning, indicated by "RL-Code-Step1200," suggesting a focus on reinforcement learning for code generation over 1200 steps. It is designed to provide robust performance in programming-related tasks.

Key Capabilities

  • Code Generation: Optimized through specific training steps to generate code effectively.
  • Qwen3 Architecture: Benefits from the foundational capabilities of the Qwen3 model family.
  • Large Context Window: Supports a context length of 32768 tokens, allowing for processing and generating longer code snippets or understanding more extensive programming contexts.

Good For

  • Software Development: Assisting developers with writing and completing code.
  • Code-centric Applications: Integration into tools that require automated code generation or understanding.
  • Research in RL for Code: Exploring the effects of reinforcement learning on code generation models.

Finetunes
1 models