A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 14, 05:30 UTC
Add more libraries to rlhf.md (#26374)
Signed-off-by: Michael Goin <mgoin64@gmail.com>
Michael Goin committed
Oct 7, 2025 at 16:59 UTC
1b86bd8e183138236415cc798f1beb3357e4f5eb
1 file changed
+13
-2
docs/training/rlhf.md
+13
-2
@@ -1,8 +1,19 @@
1
# Reinforcement Learning from Human Feedback
2
3
-Reinforcement Learning from Human Feedback (RLHF) is a technique that fine-tunes language models using human-generated preference data to align model outputs with desired behaviors.
3
+Reinforcement Learning from Human Feedback (RLHF) is a technique that fine-tunes language models using human-generated preference data to align model outputs with desired behaviors. vLLM can be used to generate the completions for RLHF.
4
5
-vLLM can be used to generate the completions for RLHF. Some ways to do this include using libraries like [TRL](https://github.com/huggingface/trl), [OpenRLHF](https://github.com/OpenRLHF/OpenRLHF), [verl](https://github.com/volcengine/verl) and [unsloth](https://github.com/unslothai/unsloth).
5
+The following open-source RL libraries use vLLM for fast rollouts (sorted alphabetically and non-exhaustive):
6
+
7
+- [Cosmos-RL](https://github.com/nvidia-cosmos/cosmos-rl)
8
+- [NeMo-RL](https://github.com/NVIDIA-NeMo/RL)
9
+- [Open Instruct](https://github.com/allenai/open-instruct)
10
+- [OpenRLHF](https://github.com/OpenRLHF/OpenRLHF)
11
+- [PipelineRL](https://github.com/ServiceNow/PipelineRL)
12
+- [Prime-RL](https://github.com/PrimeIntellect-ai/prime-rl)
13
+- [SkyRL](https://github.com/NovaSky-AI/SkyRL)
14
+- [TRL](https://github.com/huggingface/trl)
15
+- [Unsloth](https://github.com/unslothai/unsloth)
16
+- [verl](https://github.com/volcengine/verl)
17
18
See the following basic examples to get started if you don't want to use an existing library:
19