NVIDIA · kheiss-uwzoo · Aug 24, 2026
@@ -12,4 +12,4 @@ For commands, service configuration, request and response contracts, and failure
 
 - [Workflow: Agentic retrieval](workflow-agentic-retrieval.md)
 - [Semantic retrieval](vdbs.md#semantic-retrieval)
-- [Starter kits](https://github.com/NVIDIA/NeMo-Retriever/blob/main/examples/README.md)
+- [Starter kits](https://github.com/NVIDIA/NeMo-Retriever/blob/26.08/examples/README.md)
@@ -3,7 +3,7 @@
 NeMo Retriever uses different credentials depending on what you are doing:
 
 - **`NVIDIA_API_KEY`** — Authorizes HTTP calls to [NVIDIA-hosted NIMs](https://build.nvidia.com/) (for example `ai.api.nvidia.com` and `integrate.api.nvidia.com`). Obtain this key from [build.nvidia.com](https://build.nvidia.com/). Keys typically start with `nvapi-`.
-- **NGC personal key** — Used when you install the [NeMo Retriever Helm chart](https://github.com/NVIDIA/NeMo-Retriever/blob/main/nemo_retriever/helm/README.md) so the cluster can authenticate to NGC Helm repos, pull images from `nvcr.io`, and provide `NGC_API_KEY` to in-cluster NIM workloads.
+- **NGC personal key** — Used when you install the [NeMo Retriever Helm chart](https://github.com/NVIDIA/NeMo-Retriever/blob/26.08/nemo_retriever/helm/README.md) so the cluster can authenticate to NGC Helm repos, pull images from `nvcr.io`, and provide `NGC_API_KEY` to in-cluster NIM workloads.
 
 You may need one or both, for example if you deploy with Helm from NGC and also call hosted inference APIs.
 
@@ -70,7 +70,7 @@ On Windows PowerShell you can use `$env:NGC_API_KEY = "<ngc-personal-key>"`.
 
 ## Using your NGC key with Helm { #using-your-ngc-key-with-helm }
 
-Set the chart values in the [Secrets](https://github.com/NVIDIA/NeMo-Retriever/blob/main/nemo_retriever/helm/README.md#secrets) section of the Helm chart README so the chart renders `ngc-secret` and `ngc-api`:
+Set the chart values in the [Secrets](https://github.com/NVIDIA/NeMo-Retriever/blob/26.08/nemo_retriever/helm/README.md#secrets) section of the Helm chart README so the chart renders `ngc-secret` and `ngc-api`:
 
 - `ngcImagePullSecret.create` and `ngcImagePullSecret.password` create the `ngc-secret` dockerconfigjson Secret for pulls from `nvcr.io`.
 - `ngcApiSecret.create` and `ngcApiSecret.password` create the `ngc-api` Secret with `NGC_API_KEY` and `NGC_CLI_API_KEY`. The service container maps `NGC_API_KEY` and `NVIDIA_API_KEY` from the Secret `NGC_API_KEY` key when the Secret exists.
@@ -86,4 +86,4 @@ helm install retriever ./nemo_retriever/helm \
 
 Helm accepts unknown `--set` paths without error. Paths such as `imagePullSecret`, `nimApiKey`, and `nims.ngcApiKey` do not create either Secret.
 
-For defaults and additional fields, refer to [`values.yaml`](https://github.com/NVIDIA/NeMo-Retriever/blob/main/nemo_retriever/helm/values.yaml).
+For defaults and additional fields, refer to [`values.yaml`](https://github.com/NVIDIA/NeMo-Retriever/blob/26.08/nemo_retriever/helm/values.yaml).
@@ -10,7 +10,7 @@ For air-gapped or disconnected deployments, refer to [Air-gapped and disconnecte
 
 This documentation describes two ways to run [NeMo Retriever Library](overview.md) with the [parakeet-1-1b-ctc-en-us ASR NIM microservice](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`) to extract speech from audio files:
 
-- Run the NIM locally on your cluster with the [NeMo Retriever Helm chart](https://github.com/NVIDIA/NeMo-Retriever/blob/main/nemo_retriever/helm/README.md)
+- Run the NIM locally on your cluster with the [NeMo Retriever Helm chart](https://github.com/NVIDIA/NeMo-Retriever/blob/26.08/nemo_retriever/helm/README.md)
 - Use NVIDIA Cloud Functions (NVCF) endpoints for cloud-based inference
 
 Supported file types for speech extraction today:
@@ -45,7 +45,7 @@ requirements.
 
 For Kubernetes deployments with network access to package repositories, set
 `service.installFfmpeg=true` in the
-[Helm chart](https://github.com/NVIDIA/NeMo-Retriever/blob/main/nemo_retriever/helm/README.md#1-service-image)
+[Helm chart](https://github.com/NVIDIA/NeMo-Retriever/blob/26.08/nemo_retriever/helm/README.md#1-service-image)
 to install ffmpeg/ffprobe at service startup. This runtime path requires
 package-repository network egress, a writable root filesystem, and a security
 policy that allows the image's scoped sudo use. For air-gapped clusters, refer to
@@ -61,9 +61,9 @@ This pipeline enables retrieval at the speech segment level when you enable segm
 
 ## Run Parakeet on the cluster (Helm) { #run-parakeet-on-the-cluster-helm }
 
-Use the following procedure to run the NIM on your own infrastructure. Self-hosted Parakeet runs on Kubernetes through the [NeMo Retriever Helm chart](https://github.com/NVIDIA/NeMo-Retriever/blob/main/nemo_retriever/helm/README.md). Parakeet is disabled by default and is not auto-wired into the retriever service. Refer to [Optional Helm NIMs](prerequisites-support-matrix.md#optional-helm-nims-not-auto-wired-by-default) for the enablement table. GPU pinning is documented in [Parakeet ASR](https://github.com/NVIDIA/NeMo-Retriever/blob/main/nemo_retriever/helm/README.md#audio-video-parakeet).
+Use the following procedure to run the NIM on your own infrastructure. Self-hosted Parakeet runs on Kubernetes through the [NeMo Retriever Helm chart](https://github.com/NVIDIA/NeMo-Retriever/blob/26.08/nemo_retriever/helm/README.md). Parakeet is disabled by default and is not auto-wired into the retriever service. Refer to [Optional Helm NIMs](prerequisites-support-matrix.md#optional-helm-nims-not-auto-wired-by-default) for the enablement table. GPU pinning is documented in [Parakeet ASR](https://github.com/NVIDIA/NeMo-Retriever/blob/26.08/nemo_retriever/helm/README.md#audio-video-parakeet).
 
-1. Deploy or upgrade with the [NeMo Retriever Helm chart](https://github.com/NVIDIA/NeMo-Retriever/blob/main/nemo_retriever/helm/README.md). Set both of the following values, then follow [Deployment options](deployment-options.md).
+1. Deploy or upgrade with the [NeMo Retriever Helm chart](https://github.com/NVIDIA/NeMo-Retriever/blob/26.08/nemo_retriever/helm/README.md). Set both of the following values, then follow [Deployment options](deployment-options.md).
 
     ```yaml
     nimOperator:
@@ -79,7 +79,7 @@ Use the following procedure to run the NIM on your own infrastructure. Self-host
 
     Enabling only the audio NIM deploys Parakeet and leaves `audio_grpc_endpoint` set to `null`.
 
-2. If the service will process audio or video files, set `service.installFfmpeg=true` in the Helm chart when your cluster allows runtime package installation; for air-gapped clusters, refer to [Air-gapped and disconnected deployment](deployment-options.md#air-gapped-deployment) and the [Helm chart README](https://github.com/NVIDIA/NeMo-Retriever/blob/main/nemo_retriever/helm/README.md#1-service-image) for `service.image` overrides.
+2. If the service will process audio or video files, set `service.installFfmpeg=true` in the Helm chart when your cluster allows runtime package installation; for air-gapped clusters, refer to [Air-gapped and disconnected deployment](deployment-options.md#air-gapped-deployment) and the [Helm chart README](https://github.com/NVIDIA/NeMo-Retriever/blob/26.08/nemo_retriever/helm/README.md#1-service-image) for `service.image` overrides.
 
 3. After the services are running, interact with the pipeline from Python (refer to the [Python API guide](nemo-retriever-api-reference.md) for parameter details).
 

@@ -7,7 +7,7 @@ These terms appear throughout NeMo Retriever Library documentation.
 An **ingestion job** is a unit of work you run on input content (documents, audio, video, and other supported types). Submit jobs through any of these supported entry points:
 
 - **Python API** — `Ingestor` task chains such as `.extract(...)`. Library and batch modes run ingest in-process. Against a deployed Retriever service (`run_mode="service"`), the client wraps the REST contract below. Refer to the [Python API guide](nemo-retriever-api-reference.md).
-- **`retriever ingest` CLI** — including `retriever ingest service` for a running service. Refer to the [CLI reference](https://github.com/NVIDIA/NeMo-Retriever/tree/main/nemo_retriever/docs/cli).
+- **`retriever ingest` CLI** — including `retriever ingest service` for a running service. Refer to the [CLI reference](https://github.com/NVIDIA/NeMo-Retriever/tree/26.08/nemo_retriever/docs/cli).
 - **Retriever service REST API** — the public two-step ingest workflow:
   1. Create and configure the job aggregate with `POST /v1/ingest/job` and an `application/json` `JobCreateRequest` body. The JSON sets job-level fields such as `expected_documents`; it does not embed document bytes.
   2. Upload document content separately with multipart requests to job-scoped endpoints such as `POST /v1/ingest/job/{job_id}/document`.
@@ -45,7 +45,7 @@ Token-based splitting uses the revision-pinned tokenizer for the default embeddi
 ## Deployment modes { #deployment-modes }
 
 - **Library mode** — Run without the full container stack where appropriate; refer to [Deployment options](deployment-options.md).
-- **Kubernetes / Helm (self-hosted)** — Refer to [Deploy (Helm chart)](https://github.com/NVIDIA/NeMo-Retriever/blob/main/nemo_retriever/helm/README.md) and [deployment options](deployment-options.md) for running the full microservices pipeline on your infrastructure.
-- **Notebooks** — [Jupyter examples](https://github.com/NVIDIA/NeMo-Retriever/blob/main/examples/README.md) for experimentation and RAG demos.
+- **Kubernetes / Helm (self-hosted)** — Refer to [Deploy (Helm chart)](https://github.com/NVIDIA/NeMo-Retriever/blob/26.08/nemo_retriever/helm/README.md) and [deployment options](deployment-options.md) for running the full microservices pipeline on your infrastructure.
+- **Notebooks** — [Jupyter examples](https://github.com/NVIDIA/NeMo-Retriever/blob/26.08/examples/README.md) for experimentation and RAG demos.
 
 For a concise comparison, refer to [Deployment options](deployment-options.md).
@@ -289,7 +289,7 @@ The following enums are used by this schema:
 
 The following is an example JSON representation of metadata. 
 This is an example only, and does not contain the full metadata.
-For the full file, refer to the [data folder](https://github.com/NVIDIA/NeMo-Retriever/blob/main/data/multimodal_test.json).
+For the full file, refer to the [data folder](https://github.com/NVIDIA/NeMo-Retriever/blob/26.08/data/multimodal_test.json).
 
 ```json
 {

@@ -35,26 +35,26 @@ Use UDFs when you need a small, self-contained transformation that is not covere
 
 ### Repository guides
 
-- [NeMo Retriever graph README — `UDFOperator`](https://github.com/NVIDIA/NeMo-Retriever/tree/main/nemo_retriever/src/nemo_retriever/graph#using-udfoperator) — API, lifecycle, and when to use `UDFOperator` versus a custom operator class
-- [NimClient and custom NIM endpoints](https://github.com/NVIDIA/NeMo-Retriever/blob/main/nemo_retriever/developer_docs/nimclient.md#nimclient-and-custom-nim-endpoints) — call custom or self-hosted NIM microservices from UDF stages
+- [NeMo Retriever graph README — `UDFOperator`](https://github.com/NVIDIA/NeMo-Retriever/tree/26.08/nemo_retriever/src/nemo_retriever/graph#using-udfoperator) — API, lifecycle, and when to use `UDFOperator` versus a custom operator class
+- [NimClient and custom NIM endpoints](https://github.com/NVIDIA/NeMo-Retriever/blob/26.08/nemo_retriever/developer_docs/nimclient.md#nimclient-and-custom-nim-endpoints) — call custom or self-hosted NIM microservices from UDF stages
 
 ## Custom graph pipelines { #custom-graph-pipelines }
 
 When you need to compose pipelines stage-by-stage, reuse operators across workflows, or run the same graph in-process or with Ray Data, use the **graph execution model** instead of (or alongside) the fluent `GraphIngestor` API.
 
 The graph package provides `AbstractOperator`, executors (`InprocessExecutor`, `RayDataExecutor`), and operator chaining with `>>`. Built-in ingestion operators live under `nemo_retriever.operators`; you can add your own operators or UDF stages anywhere in the chain.
 
-For the full guide—including custom operator classes, executors, and graph shape constraints—refer to the [NeMo Retriever graph README](https://github.com/NVIDIA/NeMo-Retriever/tree/main/nemo_retriever/src/nemo_retriever/graph#nemo-retriever-graph).
+For the full guide—including custom operator classes, executors, and graph shape constraints—refer to the [NeMo Retriever graph README](https://github.com/NVIDIA/NeMo-Retriever/tree/26.08/nemo_retriever/src/nemo_retriever/graph#nemo-retriever-graph).
 
 ## Custom vector databases { #custom-vector-databases }
 
 The supported user path for vector storage is **[LanceDB](vdbs.md)** (`vdb_op="lancedb"`). That page covers upload, semantic retrieval, metadata filtering, and LanceDB deployment characteristics.
 
-To integrate a different vector store, implement the [`VDB`](https://github.com/NVIDIA/NeMo-Retriever/blob/main/nemo_retriever/src/nemo_retriever/common/vdb/adt_vdb.py) interface and wire it through graph [`IngestVdbOperator`](https://github.com/NVIDIA/NeMo-Retriever/blob/main/nemo_retriever/src/nemo_retriever/operators/vdb.py) / [`RetrieveVdbOperator`](https://github.com/NVIDIA/NeMo-Retriever/blob/main/nemo_retriever/src/nemo_retriever/operators/vdb.py). NVIDIA validates the first-party LanceDB operator; you are responsible for testing and maintaining other backends.
+To integrate a different vector store, implement the [`VDB`](https://github.com/NVIDIA/NeMo-Retriever/blob/26.08/nemo_retriever/src/nemo_retriever/common/vdb/adt_vdb.py) interface and wire it through graph [`IngestVdbOperator`](https://github.com/NVIDIA/NeMo-Retriever/blob/26.08/nemo_retriever/src/nemo_retriever/operators/vdb.py) / [`RetrieveVdbOperator`](https://github.com/NVIDIA/NeMo-Retriever/blob/26.08/nemo_retriever/src/nemo_retriever/operators/vdb.py). NVIDIA validates the first-party LanceDB operator; you are responsible for testing and maintaining other backends.
 
 ### Repository guides
 
-- [Vector DB package (source)](https://github.com/NVIDIA/NeMo-Retriever/tree/main/nemo_retriever/src/nemo_retriever/common/vdb) — `VDB` abstract base and LanceDB reference implementation
+- [Vector DB package (source)](https://github.com/NVIDIA/NeMo-Retriever/tree/26.08/nemo_retriever/src/nemo_retriever/common/vdb) — `VDB` abstract base and LanceDB reference implementation
 
 Partner and blueprint integrations (Elasticsearch, Pinecone, Teradata, and others) are summarized on [Vector databases — Vector database partners](vdbs.md#vector-database-partners).