Major breakthroughs in LLM research: from Hidden Reasoning to Visible Thinking
The LLM research landscape is evolving rapidly. From GPT-3 (2020) to Dual-Mode models (2025) - this timeline shows the most important breakthroughs and paradigm shifts.
Research Explorer for navigating current research.
Context is crucial. Those who know the history understand current developments better and can anticipate future trends.
OpenAI o1 & DeepSeek-R1 show: LLMs can think internally in complex ways. Reasoning is not "prompt engineering" but an emergent capability trainable via GRPO & RL.
Claude 4.5 Effort Parameter, GPT-5.1 Adaptive Thinking, Qwen3 Budget: Users explicitly control reasoning depth. Shift from hidden to user-controlled.
No longer: "prompt X for more CoT". Instead: Slider from 1-10 effort. Models themselves decide how many tokens are needed for thinking. More intuitive API.
DeepSeek DSA (Dec 2025): 60% cost reduction, 3.5x speed increase, 70% memory reduction. Sparse is no longer research - it's now standard for long contexts (1M+).
Llama 4, Gemini 3: Vision + text in the same sequence instead of separate pipes. Cross-modal reasoning better than Late Fusion. Unified architecture wins.
One model, two modes: Fast (instant) + Deep (thinking). Users choose based on task. Efficient for simple questions, capable for complex ones. Best solution.