sfutenma/dpo-qwen3_4b-cot-merged_v260302-010243
TEXT GENERATIONConcurrency Cost:1Model Size:4BQuant:BF16Ctx Length:32kPublished:Mar 1, 2026License:apache-2.0Architecture:Transformer Open Weights Warm

The sfutenma/dpo-qwen3_4b-cot-merged_v260302-010243 model is a 4 billion parameter Qwen3-based instruction-tuned language model, fine-tuned by sfutenma using Supervised Fine-Tuning (SFT) and Direct Preference Optimization (DPO). It is specifically optimized for improving structured output accuracy (JSON, YAML, XML, TOML, CSV) and enhancing reasoning capabilities through Chain-of-Thought (CoT). This model excels in tasks requiring precise structured data generation and logical reasoning.

Loading preview...