China Hot Tech @ChinaHotTech
← All posts
Robotics · Embodied AI · Open Source

RLinf v0.4: Open Embodied RL Stack From Infinigence + Tsinghua

RLinf is open-source reinforcement learning infrastructure for embodied and agentic AI, jointly from Infinigence AI (无问芯穹; also Infini-AI branding) with Tsinghua and collaborators. Tag v0.4 landed on GitHub around 20 Oct 2026; Chinese press picked it up around 28 Oct. Live repo: ~5.5k stars / ~815 forks, Apache-2.0. This is training and infra software you clone — not a robot you buy.

· ~7 min read · China Hot Tech

Light editorial graphic of RLinf v0.4: modular robot Lego blocks and RL loop in mist lavender, graphite ink, electric lime and soft clay

Illustration: mist lavender / graphite / electric lime / soft clay — modular rlinf.robotics blocks + RL loop. Specs from the v0.4 GitHub release and docs; asterisked figures are vendor / release-note claims.

Most “embodied RL” headlines still mean a paper, a closed stack, or a simulator demo that never leaves the lab. RLinf v0.4 is trying to be the opposite: a cloneable PyTorch training stack with a modular real-robot API, recipes that span sim and Franka-class hardware, and enough ecosystem hooks that overseas teams can actually try it — including as the RL post-training engine inside NVIDIA Isaac Lab v3.0.

What it is

RLinf (GitHub org/repo RLinf/RLinf) describes itself as reinforcement learning infrastructure for embodied and agentic AI. The English docs live at rlinf.readthedocs.io; Chinese release notes mirror the same v0.4 story. License is Apache-2.0. Default runtime for this release moves to PyTorch 2.11 with matching SGLang / vLLM pins — the upgrade notes say do not reuse a v0.3 venv; rebuild via requirements/install.sh or the rlinf0.4 Docker images.

Where v0.3 focused on connecting simulation to real robots, v0.4 frames three pillars: a Lego-style real-robot system, high-efficiency real-robot algorithms, and full-modality models on a heterogeneous compute base. RLinf is listed in the PyTorch Ecosystem, and Isaac Lab’s own docs document RL fine-tuning of VLAs (GR00T, OpenVLA, and related) with --rl_library rlinf.

Can overseas teams use it?

Yes, as software. Clone the Apache-2.0 repo, follow the English install / example docs, or use the Isaac Lab v3.0 RLinf path for VLA post-training on Linux. You still need GPUs (or other supported accelerators), robot access or a sim suite such as LIBERO / ManiSkill / Isaac Lab, and patience for bring-up. There is no retail robot SKU attached to this release.

RLinf v0.4 engineer summary: real-robot stack, algorithms and models, hardware matrix, usability and caveats with vendor asterisks

Three columns from the release notes — real-robot API, algorithms/models, hardware matrix — plus the practical “clone it” path and asterisked vendor benches.

Real-robot stack

The headline engineering change is rlinf.robotics: arms, dexterous hands / grippers, multi-view cameras, mobile bases, and teleoperation devices are treated as composable blocks. Hardware drivers are decoupled from task definitions and environment wrappers, so swapping a body, a config, or a cross-node layout should not force rewrites of task or policy code. Older paths under rlinf.envs.realworld / hardware robot schedulers migrate into this module and rlinf.envs.real.

Platforms called out in the release: dual-arm and single-arm Franka toolchains (PICO VR teleop, HG-DAgger, OpenPI π₀.₅ SFT/deploy, Qwen-VL reward model, LeRobot dataset tooling), plus dual-arm YAM, AgileX Piper, and SO-101. That is still a research/lab footprint — not a claim that every warehouse robot plugs in overnight — but it is unusually concrete for an open Chinese RL infra release.

Algorithms and models

v0.4 ships several real-robot-oriented algorithms. Treat the success and latency numbers as vendor / release-note results* until you reproduce them:

Model coverage expands around a PyTorch-native OpenPI (π₀ / π₀.₅) rewrite aligned operator-by-operator to the official JAX reference, plus NVIDIA Cosmos3, FastWAM, π₀-FAST, Evo-1, and MolmoAct2. RL also stretches into diffusion generation via Diffusion-NFT for SD3 and Wan2.2 (including TI2V-5B), with OCR-style rewards in the examples. Agentic / reasoning recipes (e.g. Moonlight-16B GRPO, SearchR1 FSDP) sit in the same release tree.

Hardware matrix

Beyond NVIDIA, the stack lists AMD ROCm, Huawei Ascend, Moore Threads, and newly called-out Kunlunxin, Biren, MetaX (沐曦), and Intel XPU. Ascend notes claim end-to-end compute efficiency at ~75%+ of A100* across about ten embodied models on Ascend 910B3, plus throughput gains up to about 2.78×* on some GR00T / StarVLA stacks via fused ops and prefetch overlap. Moore Threads S5000 notes describe WoVR RL closed-loop curves correlating closely with A100 on LIBERO*. All of that is first-party measurement on their configs — useful as a shopping hint for China-side clusters, not as a portable FLOP guarantee.

What changes

For overseas embodied / VLA teams, the practical delta is less “another Chinese paper” and more “a maintained training engine with an Isaac Lab on-ramp.” If you already collect demos and SFT a VLA, RLinf is one open path to PPO / actor-critic / SAC-style fine-tuning with FSDP and documented Hydra YAMLs. The modular robot API matters if you bounce between Franka, Piper-class arms, and sim without forking the whole stack. Heterogeneous accelerator support matters most for labs that must train on Ascend or ROCm rather than only A100/H100.

It does not replace a robot policy product, a teleop fleet, or a closed commercial VLA API. Bring-up cost remains real: matching Isaac Lab / GR00T pins, Ray workers, and checkpoint formats takes engineering time. Keep it distinct from related-but-different ChinaHotTech coverage: ViGAR is a world-action model for humanoids; APXInf is an edge inference engine that appears as an eval backend here — not the same story; Infinesis Taiyi is a servo/gate-driver chip from a different company.

Caveats

Bottom line: RLinf v0.4 is one of the more usable open embodied-RL infrastructures coming out of the China + Tsinghua orbit — Apache-licensed, documented in English, wired into Isaac Lab, and honest enough in its release notes that you can see which numbers are their own. Clone it if you train VLAs; skip the hype if you were hoping for a robot in a box.

Sources