(Dataset based on Pinkstack/syngen-reasoning-0.6b-dataset)

This is a 14B parameter LLM designed for synthetic grounded reasoning generation between the user prompt and the final model output, specifically for dataset modifications, but can be used for multiple use cases which require reasoning.

For example, this model allows you to turn any chat dataset into a reasoning dataset as if it was generated by DeepSeek R1 or OpenAI's GPT OSS!

A few examples are included in example1.txt, example2.txt, and example3.txt.

Sampler Settings: To avoid loops, it's best to use temp = 1.0 and default parameters for everything else.

Prompt Format

System Message

<reasoning_style>deepseek_r1</reasoning_style> # Can replace deepseek_r1 with gpt_oss
<system_prompt>Original System Prompt</system_prompt>

Prompt Message

<user>User Message Here</user>
<assistant>Assistant Final Response Here (without reasoning)</assistant>

Output Format

<think>Generated Reasoning</think>

Training Details

Base Model: Qwen/Qwen3-14B
Training Epochs: 2
Learning Rate: 2e-4
Batch Size: 32
Training Method: Full Fine-Tune (FFT)
Training Platform: Nebius Token Factory
Total Cost: $84.67 USD
Total Tokens trained: ~211M tokens
Seed: 42
As of January 1, 2026, this is the biggest model ever trained for reasoning generation!