Illustration drawn by RepoMasters from the repo’s ID and its language color. Credits
AlphaUncommon

Gen-Verse/dLLM-RL

[ICLR 2026] Official code for TraceRL: Revolutionizing post-training for Diffusion LLMs, powering the SOTA TraDo series.

Owner
Gen-Verse (organization)
Stars
Stars: 523
Forks
Forks: 45
Language
Python
License
Apache-2.0
Created
2025
Topics
code-generationdiffusion-language-modelslarge-language-modelsllm-reasoningmathmatical-reasoningreinforcement-learning-algorithmsrlhf
Tier
Alpha (250 to 699 ★) What this tier means
View on GitHub

More repos by Gen-Verse

All 13 repos by Gen-Verse

Nearby in Python

All Python repos