atomimpnsc/Qwen3-0.6B-JSON-SFT-GRPO

TEXT GENERATIONPricing:Input $0.04 / Cached $0.008 / Output $0.08Concurrent Unit Cost:1Model Size:0.8BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Sep 4, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

The atomimpnsc/Qwen3-0.6B-JSON-SFT-GRPO is a 0.8 billion parameter Qwen3 model, fine-tuned from NotoriousH2/Qwen3-0.6B-JSON-SFT, specifically optimized for JSON instruction following. This model was trained using Unsloth and Huggingface's TRL library, achieving 2x faster training. It is designed for tasks requiring structured JSON output, leveraging its 32768 token context length.

Loading preview...

Model Overview

The atomimpnsc/Qwen3-0.6B-JSON-SFT-GRPO is a 0.8 billion parameter language model based on the Qwen3 architecture. It has been specifically fine-tuned from the NotoriousH2/Qwen3-0.6B-JSON-SFT model to excel at tasks requiring structured JSON output, making it suitable for applications where precise data formatting is crucial.

Key Characteristics

  • Architecture: Qwen3-based, with 0.8 billion parameters.
  • Fine-tuning: Specialized for JSON instruction following, enhancing its ability to generate well-formed JSON responses.
  • Training Efficiency: Leveraged Unsloth and Huggingface's TRL library, resulting in a 2x speedup during the training process.
  • Context Length: Supports a substantial context window of 32768 tokens, allowing for processing longer inputs and generating more complex JSON structures.

Use Cases

This model is particularly well-suited for:

  • Structured Data Extraction: Extracting information from unstructured text into a JSON format.
  • API Interaction: Generating JSON payloads or parsing API responses.
  • Configuration Generation: Creating configuration files in JSON format based on natural language instructions.
  • Data Transformation: Converting data into a standardized JSON structure.

Finetunes
1 models