rl-framework

Tag

Cards List
#rl-framework

@samsja19: We spend a lot of time designing an elegant algorithm api in prime rl that expressive and extensible but doesn't sacrif…

X AI KOLs Following · 2026-07-06 Cached

Prime-rl adds a first-class algorithms layer with six built-in RL algorithms (GRPO, MaxRL, OPD, OPSD, SFT, ECHO), making it easier to implement custom algorithms with a single file.

0 favorites 0 likes
#rl-framework

@didier_lopes: Incredible how Z. ai literally has their RL infrastructure open source. The entire OPD post-training of GLM-5.2 took on…

X AI KOLs Following · 2026-06-19 Cached

Z. ai has open-sourced its RL infrastructure, the slime framework, which enabled efficient OPD post-training of GLM-5.2 in about two days. slime is an LLM post-training framework for RL scaling that integrates Megatron and SGLang, and has been battle-tested by frontier models like GLM, Qwen, DeepSeek, and Llama.

0 favorites 0 likes
← Back to home

Submit Feedback