Inference Scaling

An in-depth technical analysis of why reasoning models like OpenAI o1 and DeepSeek-R1 increase infrastructure costs and how to optimize your test-time compute budget.

Explore the paradigm shift from massive parameter counts to inference-time compute, where smaller models like DeepSeek-R1 and OpenAI o1-mini use 'thinking' time to outsmart giants like GPT-4.

Explore the shift from training-time scaling to inference-time scaling, comparing OpenAI o3 and DeepSeek R1, and understanding how models are learning to 'think' through RLVR.