Introduce release branch 26.03 with version 26.3.0-RC1

Update all hardcoded version references from 26.1.2 to 26.3.0-RC1
across helm charts, docker-compose, FastAPI, docs, and examples.

Made-with: Cursor

Author: jdye64

docker-compose.yaml

@@ -262,7 +262,7 @@ services:
       - audio
 
   nv-ingest-ms-runtime:
-    image: nvcr.io/nvidia/nemo-microservices/nv-ingest:26.1.2
+    image: nvcr.io/nvidia/nemo-microservices/nv-ingest:26.3.0-RC1
     shm_size: 40gb # Should be at minimum 30% of assigned memory per Ray documentation
     build:
       context: ${NV_INGEST_ROOT:-.}

docs/docs/extraction/content-metadata.md

@@ -43,7 +43,7 @@ These fields apply to all content types including text, images, and tables.
 | Subtype | The type of the content for structured data types, such as table or chart. | — |
 | Content | Content extracted from the source.  | Extracted |
 | Description | A text description of the content object. | Generated |
-| Page \# | The page \# of the content in the source. Prior to 26.1.2, this field was 0-indexed. Beginning with 26.1.2, this field is 1-indexed. | Extracted |
+| Page \# | The page \# of the content in the source. Prior to 26.3.0-RC1, this field was 0-indexed. Beginning with 26.3.0-RC1, this field is 1-indexed. | Extracted |
 | Hierarchy | The location or order of the content within the source.  | Extracted |
 
 

docs/docs/extraction/helm.md

@@ -3,4 +3,4 @@
 <!-- Use this documentation to deploy [NeMo Retriever Library](overview.md) by using Helm. -->
 
 To deploy [NeMo Retriever Library](overview.md) by using Helm, 
-refer to [NeMo Retriever Helm Charts](https://github.com/NVIDIA/NeMo-Retriever/blob/release/26.1.2/helm/README.md).
+refer to [NeMo Retriever Helm Charts](https://github.com/NVIDIA/NeMo-Retriever/blob/release/26.3.0-RC1/helm/README.md).

docs/docs/extraction/quickstart-guide.md

@@ -84,7 +84,7 @@ h. Run the command `docker ps`. You should see output similar to the following.
     CONTAINER ID  IMAGE                                            COMMAND                 CREATED         STATUS                  PORTS            NAMES
 uv venv --python 3.12 nv-ingest-dev
 source nv-ingest-dev/bin/activate
-uv pip install nv-ingest==26.1.2 nv-ingest-api==26.1.2 nv-ingest-client==26.1.2
+uv pip install nv-ingest==26.3.0-RC1 nv-ingest-api==26.3.0-RC1 nv-ingest-client==26.3.0-RC1
 ```
 
 !!! tip

docs/docs/extraction/quickstart-library-mode.md

@@ -34,7 +34,7 @@ Use the following procedure to prepare your environment.
     ```
        uv venv --python 3.12 nvingest && \
          source nvingest/bin/activate && \
-         uv pip install nemo-retriever==26.1.2 milvus-lite==2.4.12
+         uv pip install nemo-retriever==26.3.0-RC1 milvus-lite==2.4.12
     ```
 
     !!! tip

docs/docs/extraction/releasenotes-nv-ingest.md

@@ -8,19 +8,19 @@ This documentation contains the release notes for [NeMo Retriever Library](overv
 
 
 
-## Release 26.01 (26.1.2)
+## Release 26.01 (26.3.0-RC1)
 
 The NeMo Retriever Library 26.01 release adds new hardware and software support, and other improvements.
 
-To upgrade the Helm Charts for this version, refer to [NeMo Retriever Library Helm Charts](https://github.com/NVIDIA/NeMo-Retriever/blob/release/26.1.2/helm/README.md).
+To upgrade the Helm Charts for this version, refer to [NeMo Retriever Library Helm Charts](https://github.com/NVIDIA/NeMo-Retriever/blob/release/26.3.0-RC1/helm/README.md).
 
 
 ### Highlights 
 
 This release contains the following key changes:
 
 - Added functional support for [H200 NVL](https://www.nvidia.com/en-us/data-center/h200/). For details, refer to [Support Matrix](support-matrix.md).
-- All Helm deployments for Kubernetes now use [NVIDIA NIM Operator](https://docs.nvidia.com/nim-operator/latest/index.html). For details, refer to [NeMo Retriever Library Helm Charts](https://github.com/NVIDIA/NeMo-Retriever/blob/release/26.1.2/helm/README.md). 
+- All Helm deployments for Kubernetes now use [NVIDIA NIM Operator](https://docs.nvidia.com/nim-operator/latest/index.html). For details, refer to [NeMo Retriever Library Helm Charts](https://github.com/NVIDIA/NeMo-Retriever/blob/release/26.3.0-RC1/helm/README.md). 
 - Updated RIVA NIM to version 1.4.0. For details, refer to [Extract Speech](audio.md).
 - Updated VLM NIM to [nemotron-nano-12b-v2-vl](https://build.nvidia.com/nvidia/nemotron-nano-12b-v2-vl/modelcard). For details, refer to [Extract Captions from Images](python-api-reference.md#extract-captions-from-images).
 - Added VLM caption prompt customization parameters, including reasoning control. For details, refer to [Caption Images and Control Reasoning](python-api-reference.md#caption-images-and-control-reasoning).
@@ -33,7 +33,7 @@ This release contains the following key changes:
 - Large PDFs are now automatically split into chunks and processed in parallel, delivering faster ingestion for long documents. For details, refer to [PDF Pre-Splitting](v2-api-guide.md).
 - Issues maintaining extraction quality while processing very large files are now resolved with the V2 API. For details, refer to [V2 API Guide](v2-api-guide.md).
 - Updated the embedding task to support embedding on custom content fields like the results of summarization functions. For details, refer to [Use the Python API](python-api-reference.md).
-- User-defined function summarization is now using `nemotron-mini-4b-instruct` which provides significant speed improvements. For details, refer to [User-defined Functions](user-defined-functions.md) and [NeMo Retriever Library UDF Examples](https://github.com/NVIDIA/NeMo-Retriever/blob/release/26.1.2/examples/udfs/README.md).
+- User-defined function summarization is now using `nemotron-mini-4b-instruct` which provides significant speed improvements. For details, refer to [User-defined Functions](user-defined-functions.md) and [NeMo Retriever Library UDF Examples](https://github.com/NVIDIA/NeMo-Retriever/blob/release/26.3.0-RC1/examples/udfs/README.md).
 - In the `Ingestor.extract` method, the defaults for `extract_text` and `extract_images` are now set to `true` for consistency with `extract_tables` and `extract_charts`. For details, refer to [Use the Python API](python-api-reference.md).
 - The `table-structure` profile is no longer available. The table-structure profile is now part of the default profile. For details, refer to [Profile Information](quickstart-guide.md#profile-information).
 - New documentation [Why Throughput Is Dataset-Dependent](throughput-is-dataset-dependent.md).
@@ -49,8 +49,8 @@ This release contains the following key changes:
 
 The following are the known issues that are fixed in this version:
 
-- A10G support is restored. To use A10G hardware, use release 26.1.2 or later. For details, refer to [Support Matrix](support-matrix.md).
-- L40S support is restored. To use L40S hardware, use release 26.1.2 or later. For details, refer to [Support Matrix](support-matrix.md).
+- A10G support is restored. To use A10G hardware, use release 26.3.0-RC1 or later. For details, refer to [Support Matrix](support-matrix.md).
+- L40S support is restored. To use L40S hardware, use release 26.3.0-RC1 or later. For details, refer to [Support Matrix](support-matrix.md).
 - The page number field in the content metadata now starts at 1 instead of 0 so each page number is no longer off by one from what you would expect. For details, refer to [Content Metadata](content-metadata.md).
 - Support for batches that include individual files greater than approximately 400MB is restored. This includes audio files and pdfs.
 

docs/docs/extraction/user-defined-functions.md

@@ -941,6 +941,6 @@ def debug_udf(control_message: IngestControlMessage) -> IngestControlMessage:
 
 ## Related Topics
 
-- [NeMo Retriever Library UDF Examples](https://github.com/NVIDIA/NeMo-Retriever/blob/release/26.1.2/examples/udfs/README.md)
+- [NeMo Retriever Library UDF Examples](https://github.com/NVIDIA/NeMo-Retriever/blob/release/26.3.0-RC1/examples/udfs/README.md)
 - [User-Defined Stages for NeMo Retriever Library](user-defined-stages.md)
 - [NimClient Usage](nimclient.md)

examples/building_vdb_operator.ipynb

@@ -486,7 +486,7 @@
     "    self.write_to_index(records)\n",
     "```\n",
     "\n",
-    "This method is called by the NV-Ingest Ingestor class during the ingestion pipeline. For more information on how operators are integrated into NV-Ingest, refer to the [interface implementation](https://github.com/NVIDIA/nv-ingest/blob/release/26.1.2/client/src/nv_ingest_client/client/interface.py#L324).\n",
+    "This method is called by the NV-Ingest Ingestor class during the ingestion pipeline. For more information on how operators are integrated into NV-Ingest, refer to the [interface implementation](https://github.com/NVIDIA/nv-ingest/blob/release/26.3.0-RC1/client/src/nv_ingest_client/client/interface.py#L324).\n",
     "\n",
     "The simplicity of this method belies its importance - it ensures that indexes are properly configured before data ingestion begins."
    ]
@@ -728,12 +728,12 @@
     "\n",
     "This implementation includes all the features covered in this tutorial:\n",
     "\n",
-    "- ✅ Complete OpenSearch integration with k-NN vector search\n",
-    "- ✅ Configurable connection parameters and index settings\n",
-    "- ✅ Robust data validation and content filtering\n",
-    "- ✅ Efficient batch processing and error handling\n",
-    "- ✅ NVIDIA embedding model integration for query vectorization\n",
-    "- ✅ Optimized response formatting and payload management\n",
+    "- \u2705 Complete OpenSearch integration with k-NN vector search\n",
+    "- \u2705 Configurable connection parameters and index settings\n",
+    "- \u2705 Robust data validation and content filtering\n",
+    "- \u2705 Efficient batch processing and error handling\n",
+    "- \u2705 NVIDIA embedding model integration for query vectorization\n",
+    "- \u2705 Optimized response formatting and payload management\n",
     "\n",
     "### Getting Started with the OpenSearch Operator\n",
     "\n",

helm/Chart.yaml

@@ -2,7 +2,7 @@ apiVersion: v2
 name: nv-ingest
 description: NV-Ingest Microservice
 type: application
-version: 26.1.2
+version: 26.3.0-RC1
 maintainers:
   - name: NVIDIA Corporation
     url: https://www.nvidia.com/

helm/README.md

@@ -45,7 +45,7 @@ To install or upgrade the Helm chart, run the following code.
 helm upgrade \
     --install \
     nv-ingest \
-    https://helm.ngc.nvidia.com/nvidia/nemo-microservices/charts/nv-ingest-26.1.2.tgz \
+    https://helm.ngc.nvidia.com/nvidia/nemo-microservices/charts/nv-ingest-26.3.0-RC1.tgz \
     -n ${NAMESPACE} \
     --username '$oauthtoken' \
     --password "${NGC_API_KEY}" \
@@ -54,7 +54,7 @@ helm upgrade \
     --set ngcApiSecret.create=true \
     --set ngcApiSecret.password="${NGC_API_KEY}" \
     --set image.repository="nvcr.io/nvidia/nemo-microservices/nv-ingest" \
-    --set image.tag="26.1.2"
+    --set image.tag="26.3.0-RC1"
 ```
 
 Optionally you can create your own versions of the `Secrets` if you do not want to use the creation via the helm chart.
@@ -105,7 +105,7 @@ For more information, refer to [NV-Ingest-Client](https://github.com/NVIDIA/nv-i
 # Just to be cautious we remove any existing installation
 pip uninstall nv-ingest-client
 
-pip install nv-ingest-client==26.1.2
+pip install nv-ingest-client==26.3.0-RC1
 ```
 
 #### Rest Endpoint Ingress
@@ -347,7 +347,7 @@ You can also use NV-Ingest's Python client API to interact with the service runn
 | fullnameOverride | string | `""` |  |
 | image.pullPolicy | string | `"IfNotPresent"` |  |
 | image.repository | string | `"nvcr.io/nvidia/nemo-microservices/nv-ingest"` |  |
-| image.tag | string | `"26.1.2"` |  |
+| image.tag | string | `"26.3.0-RC1"` |  |
 | imagePullSecrets[0].name | string | `"ngc-api"` |  |
 | imagePullSecrets[1].name | string | `"ngc-secret"` |  |
 | ingress.annotations | object | `{}` |  |