Reasoning Models — How “Thinking” Actually Works

📰 Medium · LLM

Learn how LLMs mechanically process information and make decisions through training procedures and reward models

intermediate Published 21 May 2026
Action Steps
  1. Read about the training procedure of LLMs to understand how they learn
  2. Explore process reward models and their role in LLM decision-making
  3. Investigate GRPO and its application in LLMs
  4. Apply inference-time scaling to optimize LLM performance
  5. Analyze how these mechanical processes enable LLMs to make decisions and generate text
Who Needs to Know This

AI engineers and researchers can benefit from understanding the inner workings of LLMs to improve their models and applications

Key Insight

💡 LLMs make decisions through complex mechanical processes involving training, reward models, and scaling

Share This
🤖 Discover how LLMs 'think' through mechanical processes like training procedures and reward models

Key Takeaways

Learn how LLMs mechanically process information and make decisions through training procedures and reward models

Full Article

o1, R1, LLM Thinking — what’s actually happening mechanically. The training procedure, process reward models, GRPO, inference-time scaling… Continue reading on Medium »
Read full article → ← Back to Reads

Related Videos

5 Levels of AI Agents - From Simple LLM Calls to Multi-Agent Systems
5 Levels of AI Agents - From Simple LLM Calls to Multi-Agent Systems
Dave Ebbelaar (LLM Eng)
CREATE Your OWN Custom GPTs in ChatGPT and Gemini GEMs NOW!
CREATE Your OWN Custom GPTs in ChatGPT and Gemini GEMs NOW!
DroidCrunch
Merlin AI Review 2026: Grok, Claude, ChatGTP, Gemini - All PRO Models
Merlin AI Review 2026: Grok, Claude, ChatGTP, Gemini - All PRO Models
DroidCrunch
These 4 Gemini Features Changed How I Use Google Docs
These 4 Gemini Features Changed How I Use Google Docs
Aga Murdoch | AI Training
Notebook LLM vs PoppyAI #ai #productivity #chatgpt
Notebook LLM vs PoppyAI #ai #productivity #chatgpt
Poppy AI
NEW GPT 5.6 Models and ChatGPT Work App
NEW GPT 5.6 Models and ChatGPT Work App
Tech Friend AJ