2025 LLM Year in Review
Original excerpt
2025 has been a strong and eventful year of progress in LLMs. The following is a list of personally notable and mildly surprising "paradigm changes" - things that altered the landscape and stood out to me conceptually.
At the start of 2025, the LLM production stack in all labs looked something like this:
1. Pretraining (GPT-2/3 of ~2020) 2. Supervised Finetuning (InstructGPT ~2022) and 3. Reinforcement Learning from Human Feedback (RLHF ~2022)
This was the stable and proven recipe for training a production-grade LLM for a while. In 2025, Reinforcement Learning from Verifiable Rewards (RLVR) emerged as the de facto new major stage to add to this mix. By training LLMs against automatically…
Frequently asked questions
What is "2025 LLM Year in Review" about?
This article by Andrej Karpathy is part of the Andrej Karpathy reading list on Burn 451, covering ai agents.
Who wrote "2025 LLM Year in Review"?
This piece is part of the Andrej Karpathy vault on Burn 451, covering ai agents. The original author is attributed at the source link.
How can I read more content from Andrej Karpathy?
The complete Andrej Karpathy reading list is available at burn451.cloud/vault/karpathy. Each article includes an AI-generated summary so you can decide what to read in seconds. Connect the Burn 451 MCP server to Claude or Cursor to query all Andrej Karpathy articles as live AI context.
Can I use "2025 LLM Year in Review" with Claude or Cursor?
Yes. Install the burn-mcp-server npm package and connect it to Claude Desktop, Claude Code, or Cursor. Once connected, your AI can search and reference this article and the full Andrej Karpathy vault in real time — no manual copy-paste required.
37 more articles in this vault.
Import the full Andrej Karpathy vault to Burn 451 and build your own knowledge base.
Content attributed to the original author. Burn 451 curates publicly available writing as a reading index. For removal requests, contact @hawking520.