openrlhf-training SKILL.md skillB
openrlhf-training from orchestra-research/ai-research-skills is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add orchestra-research/ai-research-skills/openrlhf-training@v1.0.0Pins v1.0.0 in mdr.lock. Pin the fingerprint instead.
- Audit grade
- B, 16 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- orchestra-research/ai-research-skills
- Path
06-post-training/openrlhf/SKILL.md- Versions recorded
- 5
- Latest version
- v1.0.0, committed Nov 20, 2025
- First version recorded
- Nov 19, 2025
- Size
- 8,379 bytes
- Repository stars
- 13,718
- License
- MIT
High-performance RLHF framework with Ray+vLLM acceleration. Use for PPO, GRPO, RLOO, DPO training of large models (7B-70B+). Built on Ray, vLLM, ZeRO-3. 2× faster than DeepSpeedChat with distributed architecture and GPU resource sharing.
Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.
What the file says
# OpenRLHF - High-Performance RLHF Training
## Quick start
OpenRLHF is a Ray-based RLHF framework optimized for distributed training with vLLM inference acceleration.
**Installation**:
```bash
# Launch Docker container
docker run --runtime=nvidia -it --rm --shm-size="10g" --cap-add=SYS_ADMIN \
-v $PWD:/openrlhf nvcr.io/nvidia/pytorch:25.02-py3 bash
# Uninstall conflicts
sudo pip uninstall xgboost transformer_engine flash_attn pynvml -y
# Install OpenRLHF with vLLM
pip install openrlhf[vllm]
```
**PPO Training** (Hybrid Engine):
```bash
ray start --head --node-ip-address 0.0.0.0 --num-gpus 8
ray job submit --address="http://127.0.0.1:8265" \
--runtime-env-json='{"working_dir": "/openrlhf"}' \
-- python3 -m openrlhf.cli.train_ppo_ray \
--ref_num_nodes 1 --ref_num_gpus_per_node 8 \
--reward_num_nodes 1 --reward_num_gpus_per_node 8 \
--critic_num_nodes 1 --critic_num_gpus_per_node 8 \
--actor_num_nodes 1 --actor_num_gpus_per_node 8 \
--vllm_num_engines 4 --vllm_tensor_parallel_size 2 \
--colocate_all_models \
--vllm_gpu_memory_utilization 0.5 \
--pretrain OpenRLHF/Llama-3-8b-sft-mixture \
--reward_pretrain OpenRLHF/Llama-3-8b-rm-700k \
…Read the whole file at its exact version.
How to install
mdr add orchestra-research/ai-research-skills/openrlhf-training@sha256:eb9da6a1a81c492fThe line at the top of this page pins the label v1.0.0, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.
[](https://markdownregistry.com/a/art_6qil3hrnpvkjhy2u)
2 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 5 versions of this file, between Nov 19, 2025 and Nov 20, 2025, more than one a day.
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| v1.0.0 latest | 2025-11-20 | 0a9ded4 | 8,379 B | B | view · diff |
| v1.0.0 | 2025-11-20 | 72c8b2c | 8,379 B | B | view · diff |
| v1.0.0 | 2025-11-20 | 85a6d14 | 8,378 B | B | view · diff |
| v1.0.0 | 2025-11-20 | f38509b | 8,377 B | B | view · diff |
| v1.0.0 | 2025-11-19 | bd4461f | 8,376 B | B | view |
Audit of the latest version
- fail: No link to a raw IP address (matched: http://127.0.0.1:8265)
- pass: Frontmatter block present
- pass: Frontmatter declares a name
- pass: Frontmatter declares a description
- pass: Size between 200 bytes and 200 KB (8379 bytes)
- pass: No zero-width or bidi control characters
- pass: No instruction hidden inside an HTML comment
- pass: No link to an exfiltration or paste host
- pass: No credential-shaped string
- pass: No instruction to send local credentials anywhere
- pass: No text hidden with inline styles
- pass: No prompt-injection phrasing
- pass: No curl or wget piped into a shell
- pass: No recursive delete of root, home or parent
- pass: No instruction to read or print local credentials
- pass: No base64 blob over 200 characters
- pass: No script tag
Starting a new one? Use a template for SKILL.md on mdpantry, which passes every one of these checks and says what to change in it.
Source
orchestra-research/ai-research-skills · 13,718 stars · license MIT · pushed 2026-10-11 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_6qil3hrnpvkjhy2u GET https://markdownregistry.com/api/v1/resolve?ref=orchestra-research/ai-research-skills/openrlhf-training GET https://markdownregistry.com/api/v1/blob/eb9da6a1a81c492fb143cde0714f355ab042fcfe967b31bdb1fc78a5cb0e761f
Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.
Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.
More from orchestra-research/ai-research-skills
Every file in orchestra-research/ai-research-skills