ReasonFlux-Qwen3-dpo-GGUF

ReasonFlux-Qwen3-dpo-GGUF

ReasonFlux-Qwen3-dpo is a fine-tuned version of Qwen3-1.7B, specially trained on the Gen-Verse/ReasonFlux-V2-Reasoner-DPO dataset to internalize template-augmented reasoning and structured thought processes via hierarchical reinforcement learning and direct preference optimization (DPO). This model excels in scientific and mathematical domains by producing transparent, step-by-step reasoning, symbolic derivations, detailed proofs, and robust code understanding across multiple languages, while also supporting structured outputs in formats like LaTeX, Markdown, JSON, CSV, and YAML for technical workflows. Designed for efficient deployment on mid-range GPUs, research clusters, and edge environments, ReasonFlux-Qwen3-dpo enhances coherence and adaptiveness, making it an ideal tool for advanced STEM reasoning and technical research applications.

Model Files

Quants Usage

(sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)

Here is a handy graph by ikawrakow comparing some lower-quality quant types (lower is better):