slime

@slime_framework

The LLM post-training framework for RL Scaling. github.com/THUDM/slime

Joined September 2025

Following

2,240

Followers

Replies Reposts Media

Pinned
slime
@slime_framework
Aug 28
Ahead of the upcoming open-source release of GLM-5.3, we’re releasing slime v0.3.2 🚀 Highlights: - Fully aligned GLM-5 training in Megatron with SGLang rollout - Tons of community-contributed bug fixes and optimizations - Continued refactoring and codebase cleanup, including
Release v0.3.2 · THUDM/slime
From github.com
1
slime
@slime_framework
Aug 27
Unbelievable effort from the team! Glad to see slime finally helps ship a multimodal model. We’ll gradually open-source the training implementation as the corresponding support in upstream frameworks stabilizes.
Z.ai
@Zai_org
Aug 26
Introducing GLM-5.3-Flash - Leading capabilities at a highly competitive price - Natively multimodal with a 1M-token context window - A 320B-A18B model released under the MIT License - Previously previewed as Ox Alpha, running entirely on Chinese AI chips Blog:
2
slime
@slime_framework
Aug 19
Congrats 🎉🎉🎉 Amazing work by the great @radixark team!
RadixArk
@radixark
Aug 18
Today we're launching Miles v0.1, an open-source RL framework for LLMs and multimodal models. RL training is easy to start and hard to debug. Miles helps you ensure your run is correct, use hardware efficiently, and keep RL running at scale. Over the past 9 months, 72
00:00
2
slime
@slime_framework
Aug 14
Proudly presented by the team ❤️
Z.ai
@Zai_org
Aug 14
Introducing GLM-5.3: Built to Code. Ready for Cyber Defense. - Top-tier coding and agentic capabilities, achieved through post-training on the 743B base model - A major leap in cybersecurity, setting a new standard among open models Tech Blog: z.ai/blog/glm-5.3
slime
@slime_framework
Aug 12
slime ❤️ SGLang
SGLang
@sgl_project
Aug 11
Our ecosystem project and the RL framework behind GLM series training, Slime, just open-sourced its deterministic train–rollout alignment path for GLM-5.2. Megatron training and SGLang rollout matched down to a 4096-token logprob MAE of 1.9e-7, with exact zero hidden-state diff
1