Cursor Composer 2: RL Algorithm Details
Source: Tweet thread Author: Unknown (ML researcher) Date: 2026-03-30
Summary
Tweet revealing key details about Cursor Composer 2’s RL training algorithm — very close to GRPO but with specific modifications.
Key Claims
- Uses GRPO-like algorithm but removes length normalization.
- Removes group advantage standardization.
- Adds KL term with K1 variant (simpler KL approximation).
- Non-linear length penalty: penalizes very long outputs more than linearly.
- These choices suggest Cursor found standard GRPO had verbosity and instability issues that these modifications address.
Connection to Other Sources
Complements GRPO++ and Cursor Tab Online RL. Shows Cursor’s continued investment in RL for code completion.
Concepts
- RL Infrastructure — production RL details from a major coding agent
- Coding Agents — Cursor’s RL-based improvements