Fix links to custom usage from pipelines docs

Reported on discord https://discord.com/channels/879548962464493619/1142943169068154950/1142943169068154950

docs/source/pipelines.mdx

@@ -23,7 +23,7 @@ When running for the first time, the `pipeline` will download and cache the defa
 
 <Tip>
 
-By default, models will be downloaded from the [Hugging Face Hub](https://huggingface.co/models) and stored in [browser cache](https://developer.mozilla.org/en-US/docs/Web/API/Cache), but there are ways to specify custom models and cache locations. For more information see [here](/custom_usage).
+By default, models will be downloaded from the [Hugging Face Hub](https://huggingface.co/models) and stored in [browser cache](https://developer.mozilla.org/en-US/docs/Web/API/Cache), but there are ways to specify custom models and cache locations. For more information see [here](./custom_usage).
 
 </Tip>
 
@@ -52,7 +52,7 @@ let result = await reviewer('The Shawshank Redemption is a true masterpiece of c
 // [{label: '5 stars', score: 0.8167929649353027}]
 ```
 
-Transformers.js supports loading any model hosted on the Hugging Face Hub, provided it has ONNX weights (located in a subfolder called `onnx`). For more information on how to convert your PyTorch, TensorFlow, or JAX model to ONNX, see the [conversion section](/custom_usage#convert-your-models-to-onnx).
+Transformers.js supports loading any model hosted on the Hugging Face Hub, provided it has ONNX weights (located in a subfolder called `onnx`). For more information on how to convert your PyTorch, TensorFlow, or JAX model to ONNX, see the [conversion section](./custom_usage#convert-your-models-to-onnx).
 
 The `pipeline()` function is a great way to quickly use a pretrained model for inference, as it takes care of all the preprocessing and postprocessing for you. For example, if you want to perform Automatic Speech Recognition (ASR) using OpenAI's Whisper model, you can do: