Xiaomi MiMo open-sources its RL environments and training code on a fork of verl
eliebakouch · x · 2026-09-26
Xiaomi's MiMo team has open-sourced its OSS RL environments along with training code, released as XiaomiMiMo/verl — a fork of verl, the production-ready RL post-training framework for LLMs initiated by ByteDance Seed and the open-source implementation of HybridFlow. The repo ships full engineering assets including recipes, examples, docker configs, and agent skills for RL post-training experiments.
More from coding & agent
- Polylane swapped LLMs for decision model Jev in prod, cutting costs 39% — multiply_matrix · 2026-09-26
- Real-time scam call detection with ElevenLabs and Jev, ~180ms at $0.0004 — TheMoonMidas · 2026-09-26
- Real-world case: Opus 5.5 clearly beats GPT-6 Sol on architecture-level coding — DataLearnerAI · 2026-09-26
- Claude keeps killing its own grep and shell processes, and prompts don't fix it — SebastianNehrd2 · 2026-09-26
- Dev builds NVIDIA Blackwell GPU 3D animation in one HTML file with Claude Opus 5.5 — EricBuess · 2026-09-26
- Tech conference wrap-up: AI coding is evolving into the 'agentic software factory' — ahahabbak · 2026-09-26