diff --git a/CONTRIBUTING.md b/CONTRIBUTING.md
index 7e91fed058..33d0f9cd81 100644
--- a/CONTRIBUTING.md
+++ b/CONTRIBUTING.md
@@ -536,4 +536,4 @@ Portions adopted from
- [https://github.com/nv-morpheus/Morpheus/blob/branch-24.06/CONTRIBUTING.md](https://github.com/nv-morpheus/Morpheus/blob/branch-24.06/CONTRIBUTING.md)
- [https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md](https://github.com/pytorch/pytorch/blob/master/CONTRIBUTING.md)
-- [https://github.com/dask/dask/blob/master/docs/source/develop.rst](https://github.com/dask/dask/blob/master/docs/source/develop.rst)
\ No newline at end of file
+- [https://github.com/dask/dask/blob/master/docs/source/develop.rst](https://github.com/dask/dask/blob/master/docs/source/develop.rst)
diff --git a/docs/docs/extraction/air-gapped-deployment.md b/docs/docs/extraction/air-gapped-deployment.md
index 5e55b59687..92e481df1d 100644
--- a/docs/docs/extraction/air-gapped-deployment.md
+++ b/docs/docs/extraction/air-gapped-deployment.md
@@ -4,10 +4,10 @@ This guide consolidates what you need to run NeMo Retriever Library in a secured
!!! note "Source of truth for versions"
- Image repositories and default tags change between releases. Always verify pins against the `release/26.3.0` (or your exact stack) branches of:
+ Image repositories and default tags change between releases. Always verify pins against the Git branch that matches your stack (for 26.3.x self-hosted docs, use the NeMo Retriever **`26.03`** branch as the structural baseline):
- - [NeMo Retriever `docker-compose.yaml`](https://github.com/NVIDIA/NeMo-Retriever/blob/release/26.3.0/docker-compose.yaml) for self-hosted Compose (check out the **Git tag or branch that matches 26.3.0** in your environment; `main` moves forward)
- - [`helm/values.yaml`](https://github.com/NVIDIA/NeMo-Retriever/blob/release/26.3.0/helm/values.yaml) and [`helm/README.md`](https://github.com/NVIDIA/NeMo-Retriever/blob/release/26.3.0/helm/README.md) for Kubernetes / Helm
+ - [NeMo Retriever `docker-compose.yaml`](https://github.com/NVIDIA/NeMo-Retriever/blob/26.03/docker-compose.yaml) for self-hosted Compose
+ - [`helm/values.yaml`](https://github.com/NVIDIA/NeMo-Retriever/blob/26.03/helm/values.yaml) and [`helm/README.md`](https://github.com/NVIDIA/NeMo-Retriever/blob/26.03/helm/README.md) for Kubernetes / Helm
## End-to-end workflow
@@ -62,7 +62,7 @@ Again, take the exact repository and tag from your pinned `values.yaml` for 26.3
## Helm charts and packaging artifacts
-From a connected environment, download and version-control the chart archive you install, for example (see [NV-Ingest Helm README](https://github.com/NVIDIA/NeMo-Retriever/blob/release/26.3.0/helm/README.md)):
+From a connected environment, download and version-control the chart archive you install, for example (see [NV-Ingest Helm README](https://github.com/NVIDIA/NeMo-Retriever/blob/26.03/helm/README.md)):
- `nv-ingest-26.3.0.tgz` from NGC Helm (`helm pull` with NGC credentials)
@@ -90,7 +90,7 @@ Transfer tarballs instead when the enclave has no registry yet: `docker save -o
### Docker Compose
-- Override each `*_IMAGE` / `*_TAG` environment variable (see [`docker-compose.yaml`](https://github.com/NVIDIA/NeMo-Retriever/blob/release/26.3.0/docker-compose.yaml)) so every `image:` resolves to your mirror.
+- Override each `*_IMAGE` / `*_TAG` environment variable (see [`docker-compose.yaml`](https://github.com/NVIDIA/NeMo-Retriever/blob/26.03/docker-compose.yaml)) so every `image:` resolves to your mirror.
- Keep hosted API endpoints disabled: use in-stack URLs for NIMs (defaults in the compose file already prefer `http://…` service names over `https://integrate.api.nvidia.com`).
- Provide `.env` or config alongside the compose file in the enclave; never rely on pulling new images at `up` time without registry access.
diff --git a/docs/docs/extraction/audio.md b/docs/docs/extraction/audio.md
index 6e3b2b0d84..3c09b455b3 100644
--- a/docs/docs/extraction/audio.md
+++ b/docs/docs/extraction/audio.md
@@ -92,7 +92,7 @@ To generate one extracted element for each sentence-like ASR segment, include `e
!!! tip
- For more Python examples, refer to [NV-Ingest: Python Client Quick Start Guide](https://github.com/NVIDIA/nv-ingest/blob/main/client/client_examples/examples/python_client_usage.ipynb).
+ For more Python examples, refer to the [Python Client Quick Start Guide](quickstart-guide.md#ingest_python_example).
## Use NVCF Endpoints for Cloud-Based Inference
@@ -128,7 +128,7 @@ Instead of running the pipeline locally, you can use NVCF to perform inference b
!!! tip
- For more Python examples, refer to [NV-Ingest: Python Client Quick Start Guide](https://github.com/NVIDIA/nv-ingest/blob/main/client/client_examples/examples/python_client_usage.ipynb).
+ For more Python examples, refer to the [Python Client Quick Start Guide](quickstart-guide.md#ingest_python_example).
diff --git a/docs/docs/extraction/cli-reference.md b/docs/docs/extraction/cli-reference.md
index b16616f186..3bd3b2f7c5 100644
--- a/docs/docs/extraction/cli-reference.md
+++ b/docs/docs/extraction/cli-reference.md
@@ -20,7 +20,7 @@ retriever --help
!!! tip
- There is a Jupyter notebook available to help you get started with the CLI. For more information, refer to [CLI Client Quick Start Guide](https://github.com/NVIDIA/NeMo-Retriever/blob/main/client/client_examples/examples/cli_client_usage.ipynb).
+ For CLI examples, see the [CLI section of the self-hosted quickstart](quickstart-guide.md#ingest_cli_example) or the [CLI companion notebook on the `26.03` branch](https://github.com/NVIDIA/NeMo-Retriever/blob/26.03/client/client_examples/examples/cli_client_usage.ipynb).
## Parameter Reference
diff --git a/docs/docs/extraction/nemoretriever-parse.md b/docs/docs/extraction/nemoretriever-parse.md
index 51fe136edc..00cba643f7 100644
--- a/docs/docs/extraction/nemoretriever-parse.md
+++ b/docs/docs/extraction/nemoretriever-parse.md
@@ -61,7 +61,7 @@ Use the following procedure to run the NIM locally.
!!! tip
- For more Python examples, refer to [NV-Ingest: Python Client Quick Start Guide](https://github.com/NVIDIA/nv-ingest/blob/main/client/client_examples/examples/python_client_usage.ipynb).
+ For more Python examples, refer to the [Python Client Quick Start Guide](quickstart-guide.md#ingest_python_example).
## Using NVCF Endpoints for Cloud-Based Inference
@@ -102,7 +102,7 @@ Instead of running the pipeline locally, you can use NVCF to perform inference b
!!! tip
- For more Python examples, refer to [NV-Ingest: Python Client Quick Start Guide](https://github.com/NVIDIA/nv-ingest/blob/main/client/client_examples/examples/python_client_usage.ipynb).
+ For more Python examples, refer to the [Python Client Quick Start Guide](quickstart-guide.md#ingest_python_example).
diff --git a/docs/docs/extraction/notebooks.md b/docs/docs/extraction/notebooks.md
index b7748a01aa..1a4e7363cb 100644
--- a/docs/docs/extraction/notebooks.md
+++ b/docs/docs/extraction/notebooks.md
@@ -1,6 +1,6 @@
# Notebooks for NeMo Retriever Library
-To get started using [NeMo Retriever Library](overview.md), you can try one of the ready-made notebooks that are available.
+To get started using [NeMo Retriever Library](overview.md), you can try one of the guides or ready-made notebooks that are available.
!!! note
@@ -13,10 +13,10 @@ If you plan to run benchmarking or evaluation tests, you must download the [Benc
## Getting Started
-To get started with the basics, try one of the following notebooks:
+To get started with the basics, try one of the following guides or notebooks:
-- [NV-Ingest: CLI Client Quick Start Guide](https://github.com/NVIDIA/nv-ingest/blob/main/client/client_examples/examples/cli_client_usage.ipynb)
-- [NV-Ingest: Python Client Quick Start Guide](https://github.com/NVIDIA/nv-ingest/blob/main/client/client_examples/examples/python_client_usage.ipynb)
+- [CLI Client Quick Start Guide](quickstart-guide.md#ingest_cli_example) — walkthrough for `nv-ingest-cli` / `retriever` with the self-hosted stack
+- [Python Client Quick Start Guide](quickstart-guide.md#ingest_python_example) — walkthrough for the Python client with the self-hosted stack
- [How to add metadata to your documents and filter searches](https://github.com/NVIDIA/nv-ingest/blob/main/examples/metadata_and_filtered_search.ipynb)
- [How to reindex a collection](https://github.com/NVIDIA/nv-ingest/blob/main/examples/reindex_example.ipynb)
diff --git a/docs/docs/extraction/nv-ingest-python-api.md b/docs/docs/extraction/nv-ingest-python-api.md
index f0c5092bae..7a6da9d4b3 100644
--- a/docs/docs/extraction/nv-ingest-python-api.md
+++ b/docs/docs/extraction/nv-ingest-python-api.md
@@ -8,7 +8,7 @@ The [NeMo Retriever Library](overview.md) Python API provides a simple and flexi
!!! tip
- There is a Jupyter notebook available to help you get started with the Python API. For more information, refer to [Python Client Quick Start Guide](https://github.com/NVIDIA/nv-ingest/blob/main/client/client_examples/examples/python_client_usage.ipynb).
+ For examples and walkthroughs using the Python API, refer to the [Python Client Quick Start Guide](quickstart-guide.md#ingest_python_example) and [Python API Reference](python-api-reference.md).
## Summary of Key Methods
diff --git a/docs/docs/extraction/nv-ingest_cli.md b/docs/docs/extraction/nv-ingest_cli.md
index 5ab8cd095e..981e527140 100644
--- a/docs/docs/extraction/nv-ingest_cli.md
+++ b/docs/docs/extraction/nv-ingest_cli.md
@@ -17,7 +17,7 @@ nv-ingest-cli --help
!!! tip
- There is a Jupyter notebook available to help you get started with the CLI. For more information, refer to [CLI Client Quick Start Guide](https://github.com/NVIDIA/nv-ingest/blob/main/client/client_examples/examples/cli_client_usage.ipynb).
+ For CLI examples and walkthroughs, refer to the [CLI Client Quick Start Guide](quickstart-guide.md#ingest_cli_example) and [CLI Reference](cli-reference.md).
## Examples
diff --git a/docs/docs/extraction/python-api-reference.md b/docs/docs/extraction/python-api-reference.md
index 5c2749b167..02c27b6248 100644
--- a/docs/docs/extraction/python-api-reference.md
+++ b/docs/docs/extraction/python-api-reference.md
@@ -8,7 +8,7 @@ The [NeMo Retriever Library](overview.md) Python API provides a simple and flexi
!!! tip
- There is a Jupyter notebook available to help you get started with the Python API. For more information, refer to [Python Client Quick Start Guide](https://github.com/NVIDIA/NeMo-Retriever/blob/main/client/client_examples/examples/python_client_usage.ipynb).
+ For examples and walkthroughs using the Python API, refer to the [Python Client Quick Start Guide](quickstart-guide.md#ingest_python_example) and [Use the NeMo Retriever Library Python API](nv-ingest-python-api.md).
## Summary of Key Methods
diff --git a/docs/docs/extraction/quickstart-guide.md b/docs/docs/extraction/quickstart-guide.md
index e053c73c14..12e9e76218 100644
--- a/docs/docs/extraction/quickstart-guide.md
+++ b/docs/docs/extraction/quickstart-guide.md
@@ -12,7 +12,7 @@ Use the provided [docker-compose.yaml](https://github.com/NVIDIA/nv-ingest/blob/
NIM containers on their first startup can take 10-15 minutes to pull and fully load models.
-If you prefer, you can run on Kubernetes by using [our Helm chart](https://github.com/NVIDIA/nv-ingest/blob/main/helm/README.md). Also, there are [additional environment variables](environment-config.md) you can configure.
+If you prefer, you can run on Kubernetes by using [our Helm chart](https://github.com/NVIDIA/NeMo-Retriever/blob/26.03/helm/README.md). Also, there are [additional environment variables](environment-config.md) you can configure.
a. Git clone the repo:
@@ -177,10 +177,6 @@ The following examples demonstrate how to extract text, charts, tables, and imag
### In Python
-!!! tip
-
- For more Python examples, refer to [NV-Ingest: Python Client Quick Start Guide](https://github.com/NVIDIA/nv-ingest/blob/main/client/client_examples/examples/python_client_usage.ipynb).
-
```python
import logging, os, time
@@ -319,10 +315,6 @@ image_caption:[]
### Using the `nv-ingest-cli`
-!!! tip
-
- There is a Jupyter notebook available to help you get started with the CLI. For more information, refer to [CLI Client Quick Start Guide](https://github.com/NVIDIA/nv-ingest/blob/main/client/client_examples/examples/cli_client_usage.ipynb).
-
```shell
nv-ingest-cli \
@@ -474,7 +466,7 @@ For a complete NeMo Retriever Library air-gapped checklist (Compose and Kubernet
When deploying in an air-gapped environment (no internet or NGC registry access), you must pre-stage container images on a machine with network access, then transfer and load them in the isolated environment.
1. On a machine with network access: Clone the repo, authenticate with NGC (`docker login nvcr.io`), and pull all images used by your chosen profile (for example, `docker compose --profile retrieval pull`).
-2. Save images: Export the images to archives (for example, using `docker save` for each image or a script that saves all images referenced by your [docker-compose.yaml](https://github.com/NVIDIA/NeMo-Retriever/blob/main/docker-compose.yaml)).
+2. Save images: Export the images to archives (for example, using `docker save` for each image or a script that saves all images referenced by your [docker-compose.yaml](https://github.com/NVIDIA/NeMo-Retriever/blob/26.03/docker-compose.yaml)).
3. Transfer the image archives and your `docker-compose.yaml` (and `.env` if used) to the air-gapped system.
4. On the air-gapped machine: Load the images (`docker load -i `) and start the stack with the same profile (for example, `docker compose --profile retrieval up`).
@@ -536,7 +528,7 @@ docker compose \
## Specify MIG slices for NIM models
-When you deploy NeMo Retriever Library with NIM models on MIG‑enabled GPUs, MIG device slices are requested and scheduled through the `values.yaml` file for the corresponding NIM microservice. For IBM Content-Aware Storage (CAS) deployments, this allows NeMo Retriever Library NIM pods to land only on nodes that expose the desired MIG profiles [raw.githubusercontent](https://raw.githubusercontent.com/NVIDIA/NeMo-Retriever/main/helm/README.md).
+When you deploy NeMo Retriever Library with NIM models on MIG‑enabled GPUs, MIG device slices are requested and scheduled through the `values.yaml` file for the corresponding NIM microservice. For IBM Content-Aware Storage (CAS) deployments, this allows NeMo Retriever Library NIM pods to land only on nodes that expose the desired MIG profiles [raw.githubusercontent](https://raw.githubusercontent.com/NVIDIA/NeMo-Retriever/26.03/helm/README.md).
To target a specific MIG profile—for example, a 3g.20gb slice on an A100, which is a hardware-partitioned virtual GPU instance that gives your workload a fixed mid-sized share of the A100’s compute plus 20 GB of dedicated GPU memory and behaves like a smaller independent GPU—for a given NIM, configure the `resources` and `nodeSelector` under that NIM’s values path in `values.yaml`.
diff --git a/docs/docs/extraction/releasenotes-nv-ingest.md b/docs/docs/extraction/releasenotes-nv-ingest.md
index 1ca94efe36..28a7b9fb9c 100644
--- a/docs/docs/extraction/releasenotes-nv-ingest.md
+++ b/docs/docs/extraction/releasenotes-nv-ingest.md
@@ -51,4 +51,4 @@ Highlights for the 26.03 release include:
- [Prerequisites](prerequisites.md)
- [Deploy Without Containers (Library Mode)](quickstart-library-mode.md)
- [Deploy With Docker Compose (Self-Hosted)](quickstart-guide.md)
-- [Deploy With Helm](helm.md)
\ No newline at end of file
+- [Deploy With Helm](helm.md)
diff --git a/docs/docs/extraction/support-matrix.md b/docs/docs/extraction/support-matrix.md
index b95c1fb05a..341badfb92 100644
--- a/docs/docs/extraction/support-matrix.md
+++ b/docs/docs/extraction/support-matrix.md
@@ -28,7 +28,7 @@ Advanced features require additional GPU support and disk space.
This includes the following:
- Audio extraction - parakeet-1-1b-ctc-en-us — Use the [Parakeet CTC English (en-US) ASR NIM](https://docs.nvidia.com/nim/speech/latest/asr/deploy-asr-models/parakeet-ctc-en-us.html) (`nvcr.io/nim/nvidia/parakeet-1-1b-ctc-en-us`) for processing audio files. For more information, refer to [Audio Processing](audio.md).
-- Advanced visual parsing — Use [nemotron-parse](https://docs.nvidia.com/nim/vision-language-models/latest/examples/nemotron-parse/overview.html), which adds state-of-the-art text and table extraction. For more information, refer to [Advanced Visual Parsing ](nemoretriever-parse.md).
+- Advanced visual parsing — Use [nemotron-parse](https://docs.api.nvidia.com/nim/reference/nvidia-nemotron-parse), which adds state-of-the-art text and table extraction. For more information, refer to [Advanced Visual Parsing ](nemoretriever-parse.md).
- VLM — Use [nemotron-nano-12b-v2-vl](https://build.nvidia.com/nvidia/nemotron-nano-12b-v2-vl/modelcard) for experimental image captioning of unstructured images.
!!! note
diff --git a/docs/docs/extraction/user-defined-functions.md b/docs/docs/extraction/user-defined-functions.md
index 8782c42c77..e2465dcacc 100644
--- a/docs/docs/extraction/user-defined-functions.md
+++ b/docs/docs/extraction/user-defined-functions.md
@@ -940,6 +940,6 @@ def debug_udf(control_message: IngestControlMessage) -> IngestControlMessage:
## Related Topics
-- [NV-Ingest UDF Examples](https://github.com/NVIDIA/nv-ingest/blob/release/26.1.2/examples/udfs/README.md)
+- [NV-Ingest UDF Examples](https://github.com/NVIDIA/NeMo-Retriever/blob/main/examples/udfs/README.md)
- [User-Defined Stages for NeMo Retriever Library](user-defined-stages.md)
- [NimClient Usage](nimclient.md)