From 236bf9d152136f3f199f5bc23d283551c8604576 Mon Sep 17 00:00:00 2001 From: Ethan Feng Date: Sat, 9 May 2026 10:03:42 +0800 Subject: [PATCH] [Docs] Fix RLHF example links (#42073) Signed-off-by: Ethan Feng --- docs/training/async_rl.md | 2 +- docs/training/weight_transfer/ipc.md | 4 ++-- docs/training/weight_transfer/nccl.md | 6 +++--- 3 files changed, 6 insertions(+), 6 deletions(-) diff --git a/docs/training/async_rl.md b/docs/training/async_rl.md index 172466f890..d3be23fe69 100644 --- a/docs/training/async_rl.md +++ b/docs/training/async_rl.md @@ -60,4 +60,4 @@ The key insight is that requests paused with `mode="keep"` will produce tokens f ## Example -The [async RLHF example](../examples/rl/rlhf_async_new_apis.md) demonstrates this pattern with `vllm.AsyncLLMEngine`, NCCL weight transfer, and mid-flight pause/resume with validation. +The [async RLHF example](../../examples/rl/rlhf_async_new_apis.py) demonstrates this pattern with `vllm.AsyncLLMEngine`, NCCL weight transfer, and mid-flight pause/resume with validation. diff --git a/docs/training/weight_transfer/ipc.md b/docs/training/weight_transfer/ipc.md index b9c49d04c5..69e8adce66 100644 --- a/docs/training/weight_transfer/ipc.md +++ b/docs/training/weight_transfer/ipc.md @@ -83,5 +83,5 @@ See [`IPCTrainerSendWeightsArgs`](https://github.com/vllm-project/vllm/blob/main ## Examples -- [RLHF with IPC weight syncing (offline, Ray)](../../examples/rl/rlhf_ipc.md) - Colocated training and inference on a single GPU using Ray placement groups and CUDA IPC handles -- [RLHF with IPC weight syncing (online serving, HTTP)](../../examples/rl/rlhf_http_ipc.md) - Weight transfer with a vLLM HTTP server where both server and trainer share the same GPU +- [RLHF with IPC weight syncing (offline, Ray)](../../../examples/rl/rlhf_ipc.py) - Colocated training and inference on a single GPU using Ray placement groups and CUDA IPC handles +- [RLHF with IPC weight syncing (online serving, HTTP)](../../../examples/rl/rlhf_http_ipc.py) - Weight transfer with a vLLM HTTP server where both server and trainer share the same GPU diff --git a/docs/training/weight_transfer/nccl.md b/docs/training/weight_transfer/nccl.md index d5e0f675ed..bfde1ee2ae 100644 --- a/docs/training/weight_transfer/nccl.md +++ b/docs/training/weight_transfer/nccl.md @@ -114,6 +114,6 @@ The `names`, `dtype_names`, and `shapes` lists describe each parameter. These mu ## Examples -- [RLHF with NCCL weight syncing (offline, Ray)](../../examples/rl/rlhf_nccl.md) - Trainer on one GPU, 2x tensor-parallel vLLM engine on two others, with packed NCCL weight broadcast -- [RLHF with async weight syncing (offline, Ray)](../../examples/rl/rlhf_async_new_apis.md) - Async generation with mid-flight pause, weight sync, resume, and validation against a fresh model -- [RLHF with NCCL weight syncing (online serving, HTTP)](../../examples/rl/rlhf_http_nccl.md) - Weight transfer with a running vLLM HTTP server using HTTP control plane and NCCL data plane +- [RLHF with NCCL weight syncing (offline, Ray)](../../../examples/rl/rlhf_nccl.py) - Trainer on one GPU, 2x tensor-parallel vLLM engine on two others, with packed NCCL weight broadcast +- [RLHF with async weight syncing (offline, Ray)](../../../examples/rl/rlhf_async_new_apis.py) - Async generation with mid-flight pause, weight sync, resume, and validation against a fresh model +- [RLHF with NCCL weight syncing (online serving, HTTP)](../../../examples/rl/rlhf_http_nccl.py) - Weight transfer with a running vLLM HTTP server using HTTP control plane and NCCL data plane