A LITTLE SLIME. A LOT OF POSSIBILITY.

Built for
scaling RL.

vime is an LLM post-training framework for RL scaling, providing two core capabilities:

  1. High-Performance Training ↘
  2. Flexible Data Generation ↘

Explore in your browser. Ready to run? Install vime →

01 / ↔

High-Performance Training

Supports efficient training in various modes by connecting Megatron with vLLM.

Megatron⇄vLLM
02 / ⌁

Flexible Data Generation

Enables arbitrary training data generation workflows through custom data generation interfaces and server-based engines.

Your data & environments→Data Buffer

Training, rollout, custom generation, rewards, verifiers, and environment interaction flow through the same training / rollout / Data Buffer path.

BUILT IN THE OPEN. USED AT SCALE.

From GLM-4.5 to GLM-5.3 & GLM-5.3-Flash.

slime has powered RL post-training across Z.ai’s GLM generations. Explore the model releases, then use the same core capabilities in your own experiment.

Your environment first. Your experiment next. ↓

00 / SET UP ONCE

Ready, set, vime.

Start with the prepared image. Then choose a model below and take your launch script with you.

  1. 01Install the environment
  2. 02Build your experiment
  3. 03Download, prepare, run
Environment details & other platforms →

Docker · A prepared stack

The image includes Megatron, vLLM and vime’s patches. On your Linux x86_64 GPU host, have Git, Docker and the NVIDIA driver ready, with NVIDIA Container Toolkit ↗.

Hopper / Blackwell · CUDA 13.0

1 / On the host
VIME_IMAGE=vllm/vime:latest
docker pull "$VIME_IMAGE"
git clone https://github.com/vllm-project/vime.git
cd vime
mkdir -p ../vime-data
docker run --rm -it --gpus all --ipc=host \
  --shm-size=16g --ulimit memlock=-1 --ulimit stack=67108864 \
  -v "$PWD:/root/vime" -v "$PWD/../vime-data:/data" \
  -w /root/vime "$VIME_IMAGE" /bin/bash
2 / Inside the container
pip install -e . --no-deps
python -c "import vime, vllm, megatron.core; import torch; print('GPUs:', torch.cuda.device_count())"

Your checkout stays on the host. Models, data and checkpoints under /data persist in the sibling vime-data directory after the container exits. The builder uses /data by default.

01 / THE EXPERIMENT BUILDER

Make it yours.

A starting point?
Compare A / B

Save a complete snapshot as A; your working experiment is B. Undo / Redo changes B. Replacing A is a separate action. Exports below always use B.

03 / FOLLOW YOUR CURIOSITY

Go a little deeper.

The ideas, equations, and engineering behind your choices.

01The experiment field guideFrom checkpoint to first reward ↗ 02Inside the algorithmsGRPO · TIS · ICE-POP · R3 · SC ↗ 03Inside the systemPrecision, async, PD & HiCache ↗ 04The documentation libraryOrganized guides, APIs & debugging ↗

04 / WHEN YOU WANT TO GO FURTHER

A good experiment asks good questions.

Open a question. Find the next measurement, then the right tool.

01 / DEBUGThe reward looks wrong. Where do I start?+

Inspect generated text, checkpoint loading, and the chat template first. Save a rollout, replay a fixed batch on the training side, then compare log-probs before changing the algorithm.

Debug & replay →Understand mismatch →
02 / PROFILETraining is slow. What am I waiting for?+

Separate rollout, reward, weight transfer, and training time. Trace the critical path, then profile the busy engine. Use tools/profile_rollout.py to collect vLLM traces and inspect prefill/decode behavior.

Profiling →Timeline & tracing →Metrics →
03 / SCALEHow do I grow beyond one cluster?+

Split GPU ownership, then choose a weight transport that matches the network. Add PD when prefill and decode need different resources; use persistent rollout storage when jobs need recovery.

External engines →Delta sync →Partial / distributed async →straw →Recovery →
04 / EXTENDMy agent needs its own world.+

Start with a custom per-sample generation function and reward. Replace rollout orchestration only when your data scheduling or cross-step state needs it.

Agent RL →Extension interfaces →vLLM topology →

LEARN WITHOUT LOSING YOUR PLACE