Upgrade to Nemotron RC NIMs and NV-Ingest 26.1.2 by smasurekar · Pull Request #371 · NVIDIA-AI-Blueprints/rag

smasurekar · 2026-02-19T15:10:41Z

Upgrade to Nemotron RC NIMs and NV-Ingest 26.1.2

Summary

NIMs: Switched from nemoretriever to nemotron; release candidate (RC) images are used.
NV Ingest: Upgraded to 26.1.2 (from 26.1.1).
⚠️ Cloud endpoints: Not available for these new models yet. Flows using NVIDIA-hosted (cloud) endpoints for embedding, reranker, page-elements, graphic-elements and table-structure will fail until the models are GA; they will work once generally available, with no further blueprint changes.
nvdev.env: Unchanged - new models are not on nvdev. nvdev.env is kept unchanged for CI to pass.
Nemotron OCR: Not included; has known issues. Will be added in a follow-up once resolved.

NIM / model changes

Component	Before model name	After model name
Embedding	`nvidia/llama-3.2-nv-embedqa-1b-v2`	`nvidia/llama-nemotron-embed-1b-v2`
Reranker	`nvidia/llama-3.2-nv-rerankqa-1b-v2`	`nvidia/llama-nemotron-rerank-1b-v2`
Vision / layout	`nemoretriever-page-elements-v3`, `nemoretriever-graphic-elements-v1`, `nemoretriever-table-structure-v1`	`nemotron-page-elements-v3`, `nemotron-graphic-elements-v1`, `nemotron-table-structure-v1`

Signed-off-by: Swapnil Masurekar <smasurekar@nvidia.com>

* fix(observability): use hardcoded GenAI attribute keys instead of deprecated SpanAttributes (#377) Signed-off-by: Swapnil Masurekar <smasurekar@nvidia.com> * Added MIG Slice Support for rtx6000pro (#379) * Added MIG Slice support for RTX 6000 pro Signed-off-by: Punit Kumar <punitk@nvidia.com> * Changed to default config in MIG slicing in rtx6000pro config --------- Signed-off-by: Punit Kumar <punitk@nvidia.com> Co-authored-by: niyatisingal <nsingal@nvidia.com> * Port release-v2.4.0 to release-v2.5.0 and update container versions (#385) * changes to docs per bug 5767861 (#328) * Updated launchable with v2.4.0 tag (#318) * updated support matrix (#321) * Document the end‑to‑end flow from query to answer and show how to measure time spent in each stage of the RAG pipeline. (#317) * adding oberservablility * Update docs/debugging.md Co-authored-by: nkmcalli <nkmcalli@yahoo.com> * Update docs/observability.md Co-authored-by: nkmcalli <nkmcalli@yahoo.com> * Add query-to-answer-pipeline doc and observability/debugging updates * Trigger CI * getting build to kick in for observability file * Fix typos in query-to-answer-pipeline.md and ensure file in PR for link check * get rid of PULL_REQUEST_SUMMARY --------- Co-authored-by: nkmcalli <nkmcalli@yahoo.com> * fixed files associated with build (#322) * Add multimodal query integration tests to CI pipeline * changes to docs per bug 5767861 * updated files per bug 5880717 (#327) * updated files per bug 5880717 * Update CONTRIBUTING.md * Update README.md * Update python-client.md * Update readme.md * Update readme.md * Update docs/deploy-helm.md Co-authored-by: nkmcalli <nkmcalli@yahoo.com> * Update docs/deploy-helm.md Co-authored-by: nkmcalli <nkmcalli@yahoo.com> --------- Co-authored-by: rkharwar-nv <rkharwar@nvidia.com> Co-authored-by: nkmcalli <nkmcalli@yahoo.com> Co-authored-by: Pranjal Doshi <pranjald@nvidia.com> Co-authored-by: nv-pranjald <150428320+nv-pranjald@users.noreply.github.com> * Fix workflow rule and doc bugs (#331) * Revert back milvus version in conf.md to v2.6.5 * Modify workflow to run on any branch * Fix workflow push rule to run on protected branches * Add files via upload (#326) Found an error in the Q&A section where images in the citation were not being printed. * Doc bug fixes (#339) * updated helm instructions (#333) * updated helm instructions * Update deploy-helm.md * fix broken image link (#334) * Add release note for Audio model deployment on Kubernetes on RTX‑6000 Pro is not supported in this release.heiss/5863956a (#335) * Add release note for Audio model deployment on Kubernetes on RTX‑6000 Pro is not supported in this release. * Add release note for Audio model deployment on Kubernetes on RTX‑6000 Pro is not supported in this release. * Fix broken image link in observability file * Fix CPU seach with GPU index doc * Fix VLLM profile instruction for nemotron-3-nano --------- Co-authored-by: Kurt Heiss <kheiss@nvidia.com> * Updated troubleshoot documentation for Elasticsearch connection timeout (#341) Signed-off-by: Swapnil Masurekar <smasurekar@nvidia.com> * updated path to image files so that html output is rendered correctly (#363) * Updated helm instructions for mig-deployment prerequisites (#364) * Updated helm instructions for mig-deployment * Update mig-deployment.md * Doc enhancement for noteboook (#361) * Doc enhancement for noteboook * Update release notes * Update launchable.ipynb (#365) Updated branch name State name changed from "FAILURE"->"FAILED" * Fix typo in release notes --------- Co-authored-by: rkharwar-nv <rkharwar@nvidia.com> * fixed links in deploy-helm and mig-deploymnent (#367) * update artifacts to GA version for v2.4.0 release (#359) * updated files according to style guide (#369) * Revert deploy-helm and mig-deployment to pre-11a31a4 versions (#372) * Fix release date in changelog (#373) * Bump up version to 2.5.0 --------- Signed-off-by: Swapnil Masurekar <smasurekar@nvidia.com> Co-authored-by: Kurt Heiss <kheiss@nvidia.com> Co-authored-by: rkharwar-nv <rkharwar@nvidia.com> Co-authored-by: nkmcalli <nkmcalli@yahoo.com> Co-authored-by: Pranjal Doshi <pranjald@nvidia.com> Co-authored-by: nv-pranjald <150428320+nv-pranjald@users.noreply.github.com> Co-authored-by: Swapnil Masurekar <smasurekar@nvidia.com> * feat: Event-Driven Document/Video Ingestion Pipeline (#351) * feat: add rag_event_ingest example - event-driven document/video ingestion pipeline - Kafka consumer that monitors MinIO object storage for new uploads - Routes documents to RAG Ingestor, videos to VSS for analysis - Docker Compose deployment for Kafka, MinIO, and consumer - Jupyter notebook for end-to-end deployment and testing - Sample test data (PDF document, MP4 video) tracked via Git LFS Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor * fix: polish rag_event_ingest notebook - fix sections, descriptions, TOC Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor * refactor: consolidate Setup into single cell - clone, deps, API keys Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor * refactor: inline check_rag/vss/aidp_status into their usage cells Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor * docs: add markdown description before every code cell Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor * fix: notebook is standalone entry point, clones RAG repo to ~/rag Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor * fix: set COLLECTION_NAME, load .env, simplify query_rag, add expected logs Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor * fix: left-align markdown tables in notebook Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor * fix: use HTML tables to force left alignment Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor * fix: replace API Keys markdown table with HTML for left alignment Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor * feat: add storage verification, RAG frontend hints, and configurable consumer prompts - Add verify_file_in_storage() helper to confirm files landed in MinIO - Merge storage verification into document/video ingestion checks - Add RAG Frontend UI link (port 8090) to query sections - Make Kafka consumer VSS prompts configurable via env vars in docker-compose - Install git/git-lfs in notebook setup cell - Index cells in Deploy Continuous Ingestion section Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor * fix: skip RAG clone if directory already exists Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor * fix: url Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor * feat: add continuous ingestion notebook for video and document pipeline Add rag_event_ingest.ipynb notebook that provides an end-to-end walkthrough for: - Deploying NVIDIA RAG stack (NIMs, Milvus, Ingestor, RAG Server) - Deploying NVIDIA VSS stack (VLM, LLM, Embedding, Reranker NIMs) - Deploying continuous ingestion pipeline (Kafka, MinIO, Kafka Consumer) - Configurable video analysis prompts for the Kafka consumer - Uploading documents and videos to MinIO with storage verification - Verifying ingestion via consumer logs - Querying ingested content via RAG API or Frontend UI Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor * fix: gpu Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor * fix: add ensurepip, fix VSS tag to v2.4.1, use GPUs 2-3 for VSS, update hw req to 4 GPUs Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor * fix: tag Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor * feat: resolve comment * fix: patch VSS config to use host-mapped ports for shared RAG embedding/reranker The via-server runs on the local_deployment_single_gpu_default network, not nvidia-rag, so it cannot resolve nemoretriever-embedding-ms or nemoretriever-ranking-ms. Route through host.docker.internal with the correct host-mapped ports instead (9080 for embedding, 1976 for reranker). Co-authored-by: Cursor <cursoragent@cursor.com> Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor --------- Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Co-authored-by: anngu <anngu@nvidia.com> Co-authored-by: Cursor <cursoragent@cursor.com> * Fix query decomp doc and prompt (#316) * Fix query decomp doc and prompt * fix prompt in helm as well * Upgrade to Nemotron RC NIMs and NV-Ingest 26.1.2 (#371) Signed-off-by: Swapnil Masurekar <smasurekar@nvidia.com> * Preserve filename case in filter; fix syntax error examples to use double quotes (#337) * conf.py fix (#391) * confirming presence of switcher text in conf.py file * docs: adjust conf.py for 2.5.0 * Update prompt and unify reasoning budged and enable thinking (#386) * Prompt tuining, low reasoning and reasoning budget * Filter out think token when enable filter is on * Use default prompt * Fix unit test * Add doc for nemotron thinking budget * Add question back in prompt.yaml * Concatenate multimodal content for VLM Embed (#362) (#392) Signed-off-by: Swapnil Masurekar <smasurekar@nvidia.com> * security: Fix frontend CVEs (#396) * Add config to enable nemotron parse only extraction in nv-ingest (#395) * Add config to enable nemotron parse only extraction in nv-ingest * Refactor nemotron parse only documentation * Remove nemotron parse only references from the previous section * Added patch command in rtx6000pro mig block also in documentation (#398) * Upgraded to GA nemotron-ranking-ms and nemotron-embedding-ms containers (#402) Signed-off-by: Swapnil Masurekar <smasurekar@nvidia.com> * Update packages to resolve source code CVEs (#400) * Update langchain-nvidia-ai-endpointsto >=1.1.0 * security: Update langgraph to version 1.10.0 * Update NIM wait times and patch VSS embed/rerank models (#397) * Update NIM wait times and patch VSS embed/rerank models Adjust expected NIM model loading wait from 2-5 min to ~10 min for RTX PRO 6000 hardware. Add explicit patching of VSS config.yaml to align embedding and reranker model names with RAG stack defaults. Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor * Update VSS prompts to match default format and use seconds-based queries - Align consumer VSS prompts with VSS config.yaml defaults (sports-adapted): caption, caption_summarization, summary_aggregation with proper dedup/merge logic - Extract RAG embed/rerank model names dynamically from compose file - Add parse_compose_default helper to avoid hardcoded model names - Change time-range query from MM:SS to seconds format for VSS compatibility Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor * Clean up notebook: rename variable, remove config overrides - Rename _rag_compose to _rag_compose_path for clarity - Remove hardcoded max_tokens and batch_size patches from VSS config - Simplify time-range query cell comments Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor * Rename helper function and revert prompts to MM:SS format - Rename parse_compose_default to extract_rag_default_compose_var - Revert VSS prompts to MM:SS timestamp conversion style - Revert time-range query to MM:SS format Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor * Add GPU assignment table and update NIM container names Update notebook to reflect renamed NIM containers (nemoretriever-* → nemotron-*) and add default GPU assignment table for RTX PRO 6000 / H100 hardware. Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor --------- Signed-off-by: Minh Nguyen <minhngu@nvidia.com> * docs: Add release note for v2.5.0 release (#401) Co-authored-by: Kurt Heiss <kheiss@nvidia.com> * Upgraded to GA containers for page-elements, graphic-elements, table-structure (#403) Signed-off-by: Swapnil Masurekar <smasurekar@nvidia.com> * Add SHM size 16gb to reranking and VLM NIM in compose (#405) * Fixed the query to be derived from messages if query is not explicitly given in argument and adding messages list in logs (#404) * bug: Fix frontend image path --------- Signed-off-by: Swapnil Masurekar <smasurekar@nvidia.com> Signed-off-by: Punit Kumar <punitk@nvidia.com> Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Co-authored-by: Swapnil Masurekar <smasurekar@nvidia.com> Co-authored-by: kumar-punit <punitk@nvidia.com> Co-authored-by: niyatisingal <nsingal@nvidia.com> Co-authored-by: Kurt Heiss <kheiss@nvidia.com> Co-authored-by: rkharwar-nv <rkharwar@nvidia.com> Co-authored-by: nkmcalli <nkmcalli@yahoo.com> Co-authored-by: Pranjal Doshi <pranjald@nvidia.com> Co-authored-by: nv-pranjald <150428320+nv-pranjald@users.noreply.github.com> Co-authored-by: Minh Nguyen <minhngu@nvidia.com> Co-authored-by: anngu <anngu@nvidia.com> Co-authored-by: Cursor <cursoragent@cursor.com> Co-authored-by: Nikhil Kulkarni <nikkulkarni@nvidia.com>

* fix(observability): use hardcoded GenAI attribute keys instead of deprecated SpanAttributes (#377) Signed-off-by: Swapnil Masurekar <smasurekar@nvidia.com> * Fixing nemoguardrails config (#382) Signed-off-by: Niyati Singal <nsingal@nvidia.com> * Added MIG Slice Support for rtx6000pro (#379) * Added MIG Slice support for RTX 6000 pro Signed-off-by: Punit Kumar <punitk@nvidia.com> * Changed to default config in MIG slicing in rtx6000pro config --------- Signed-off-by: Punit Kumar <punitk@nvidia.com> Co-authored-by: niyatisingal <nsingal@nvidia.com> * Port release-v2.4.0 to release-v2.5.0 and update container versions (#385) * changes to docs per bug 5767861 (#328) * Updated launchable with v2.4.0 tag (#318) * updated support matrix (#321) * Document the end‑to‑end flow from query to answer and show how to measure time spent in each stage of the RAG pipeline. (#317) * adding oberservablility * Update docs/debugging.md Co-authored-by: nkmcalli <nkmcalli@yahoo.com> * Update docs/observability.md Co-authored-by: nkmcalli <nkmcalli@yahoo.com> * Add query-to-answer-pipeline doc and observability/debugging updates * Trigger CI * getting build to kick in for observability file * Fix typos in query-to-answer-pipeline.md and ensure file in PR for link check * get rid of PULL_REQUEST_SUMMARY --------- Co-authored-by: nkmcalli <nkmcalli@yahoo.com> * fixed files associated with build (#322) * Add multimodal query integration tests to CI pipeline * changes to docs per bug 5767861 * updated files per bug 5880717 (#327) * updated files per bug 5880717 * Update CONTRIBUTING.md * Update README.md * Update python-client.md * Update readme.md * Update readme.md * Update docs/deploy-helm.md Co-authored-by: nkmcalli <nkmcalli@yahoo.com> * Update docs/deploy-helm.md Co-authored-by: nkmcalli <nkmcalli@yahoo.com> --------- Co-authored-by: rkharwar-nv <rkharwar@nvidia.com> Co-authored-by: nkmcalli <nkmcalli@yahoo.com> Co-authored-by: Pranjal Doshi <pranjald@nvidia.com> Co-authored-by: nv-pranjald <150428320+nv-pranjald@users.noreply.github.com> * Fix workflow rule and doc bugs (#331) * Revert back milvus version in conf.md to v2.6.5 * Modify workflow to run on any branch * Fix workflow push rule to run on protected branches * Add files via upload (#326) Found an error in the Q&A section where images in the citation were not being printed. * Doc bug fixes (#339) * updated helm instructions (#333) * updated helm instructions * Update deploy-helm.md * fix broken image link (#334) * Add release note for Audio model deployment on Kubernetes on RTX‑6000 Pro is not supported in this release.heiss/5863956a (#335) * Add release note for Audio model deployment on Kubernetes on RTX‑6000 Pro is not supported in this release. * Add release note for Audio model deployment on Kubernetes on RTX‑6000 Pro is not supported in this release. * Fix broken image link in observability file * Fix CPU seach with GPU index doc * Fix VLLM profile instruction for nemotron-3-nano --------- Co-authored-by: Kurt Heiss <kheiss@nvidia.com> * Updated troubleshoot documentation for Elasticsearch connection timeout (#341) Signed-off-by: Swapnil Masurekar <smasurekar@nvidia.com> * updated path to image files so that html output is rendered correctly (#363) * Updated helm instructions for mig-deployment prerequisites (#364) * Updated helm instructions for mig-deployment * Update mig-deployment.md * Doc enhancement for noteboook (#361) * Doc enhancement for noteboook * Update release notes * Update launchable.ipynb (#365) Updated branch name State name changed from "FAILURE"->"FAILED" * Fix typo in release notes --------- Co-authored-by: rkharwar-nv <rkharwar@nvidia.com> * fixed links in deploy-helm and mig-deploymnent (#367) * update artifacts to GA version for v2.4.0 release (#359) * updated files according to style guide (#369) * Revert deploy-helm and mig-deployment to pre-11a31a4 versions (#372) * Fix release date in changelog (#373) * Bump up version to 2.5.0 --------- Signed-off-by: Swapnil Masurekar <smasurekar@nvidia.com> Co-authored-by: Kurt Heiss <kheiss@nvidia.com> Co-authored-by: rkharwar-nv <rkharwar@nvidia.com> Co-authored-by: nkmcalli <nkmcalli@yahoo.com> Co-authored-by: Pranjal Doshi <pranjald@nvidia.com> Co-authored-by: nv-pranjald <150428320+nv-pranjald@users.noreply.github.com> Co-authored-by: Swapnil Masurekar <smasurekar@nvidia.com> * feat: Event-Driven Document/Video Ingestion Pipeline (#351) * feat: add rag_event_ingest example - event-driven document/video ingestion pipeline - Kafka consumer that monitors MinIO object storage for new uploads - Routes documents to RAG Ingestor, videos to VSS for analysis - Docker Compose deployment for Kafka, MinIO, and consumer - Jupyter notebook for end-to-end deployment and testing - Sample test data (PDF document, MP4 video) tracked via Git LFS Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor * fix: polish rag_event_ingest notebook - fix sections, descriptions, TOC Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor * refactor: consolidate Setup into single cell - clone, deps, API keys Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor * refactor: inline check_rag/vss/aidp_status into their usage cells Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor * docs: add markdown description before every code cell Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor * fix: notebook is standalone entry point, clones RAG repo to ~/rag Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor * fix: set COLLECTION_NAME, load .env, simplify query_rag, add expected logs Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor * fix: left-align markdown tables in notebook Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor * fix: use HTML tables to force left alignment Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor * fix: replace API Keys markdown table with HTML for left alignment Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor * feat: add storage verification, RAG frontend hints, and configurable consumer prompts - Add verify_file_in_storage() helper to confirm files landed in MinIO - Merge storage verification into document/video ingestion checks - Add RAG Frontend UI link (port 8090) to query sections - Make Kafka consumer VSS prompts configurable via env vars in docker-compose - Install git/git-lfs in notebook setup cell - Index cells in Deploy Continuous Ingestion section Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor * fix: skip RAG clone if directory already exists Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor * fix: url Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor * feat: add continuous ingestion notebook for video and document pipeline Add rag_event_ingest.ipynb notebook that provides an end-to-end walkthrough for: - Deploying NVIDIA RAG stack (NIMs, Milvus, Ingestor, RAG Server) - Deploying NVIDIA VSS stack (VLM, LLM, Embedding, Reranker NIMs) - Deploying continuous ingestion pipeline (Kafka, MinIO, Kafka Consumer) - Configurable video analysis prompts for the Kafka consumer - Uploading documents and videos to MinIO with storage verification - Verifying ingestion via consumer logs - Querying ingested content via RAG API or Frontend UI Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor * fix: gpu Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor * fix: add ensurepip, fix VSS tag to v2.4.1, use GPUs 2-3 for VSS, update hw req to 4 GPUs Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor * fix: tag Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor * feat: resolve comment * fix: patch VSS config to use host-mapped ports for shared RAG embedding/reranker The via-server runs on the local_deployment_single_gpu_default network, not nvidia-rag, so it cannot resolve nemoretriever-embedding-ms or nemoretriever-ranking-ms. Route through host.docker.internal with the correct host-mapped ports instead (9080 for embedding, 1976 for reranker). Co-authored-by: Cursor <cursoragent@cursor.com> Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor --------- Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Co-authored-by: anngu <anngu@nvidia.com> Co-authored-by: Cursor <cursoragent@cursor.com> * Fix query decomp doc and prompt (#316) * Fix query decomp doc and prompt * fix prompt in helm as well * Upgrade to Nemotron RC NIMs and NV-Ingest 26.1.2 (#371) Signed-off-by: Swapnil Masurekar <smasurekar@nvidia.com> * Preserve filename case in filter; fix syntax error examples to use double quotes (#337) * conf.py fix (#391) * confirming presence of switcher text in conf.py file * docs: adjust conf.py for 2.5.0 * Added vdb serialization if parallel ingestion which helps in high concurrent batch ingestion having indexing issues (#389) * Update prompt and unify reasoning budged and enable thinking (#386) * Prompt tuining, low reasoning and reasoning budget * Filter out think token when enable filter is on * Use default prompt * Fix unit test * Add doc for nemotron thinking budget * Add question back in prompt.yaml * Concatenate multimodal content for VLM Embed (#362) (#392) Signed-off-by: Swapnil Masurekar <smasurekar@nvidia.com> * security: Fix frontend CVEs (#396) * Add config to enable nemotron parse only extraction in nv-ingest (#395) * Add config to enable nemotron parse only extraction in nv-ingest * Refactor nemotron parse only documentation * Remove nemotron parse only references from the previous section * Added patch command in rtx6000pro mig block also in documentation (#398) * Upgraded to GA nemotron-ranking-ms and nemotron-embedding-ms containers (#402) Signed-off-by: Swapnil Masurekar <smasurekar@nvidia.com> * Update packages to resolve source code CVEs (#400) * Update langchain-nvidia-ai-endpointsto >=1.1.0 * security: Update langgraph to version 1.10.0 * Update NIM wait times and patch VSS embed/rerank models (#397) * Update NIM wait times and patch VSS embed/rerank models Adjust expected NIM model loading wait from 2-5 min to ~10 min for RTX PRO 6000 hardware. Add explicit patching of VSS config.yaml to align embedding and reranker model names with RAG stack defaults. Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor * Update VSS prompts to match default format and use seconds-based queries - Align consumer VSS prompts with VSS config.yaml defaults (sports-adapted): caption, caption_summarization, summary_aggregation with proper dedup/merge logic - Extract RAG embed/rerank model names dynamically from compose file - Add parse_compose_default helper to avoid hardcoded model names - Change time-range query from MM:SS to seconds format for VSS compatibility Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor * Clean up notebook: rename variable, remove config overrides - Rename _rag_compose to _rag_compose_path for clarity - Remove hardcoded max_tokens and batch_size patches from VSS config - Simplify time-range query cell comments Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor * Rename helper function and revert prompts to MM:SS format - Rename parse_compose_default to extract_rag_default_compose_var - Revert VSS prompts to MM:SS timestamp conversion style - Revert time-range query to MM:SS format Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor * Add GPU assignment table and update NIM container names Update notebook to reflect renamed NIM containers (nemoretriever-* → nemotron-*) and add default GPU assignment table for RTX PRO 6000 / H100 hardware. Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor --------- Signed-off-by: Minh Nguyen <minhngu@nvidia.com> * docs: Add release note for v2.5.0 release (#401) Co-authored-by: Kurt Heiss <kheiss@nvidia.com> * Upgraded to GA containers for page-elements, graphic-elements, table-structure (#403) Signed-off-by: Swapnil Masurekar <smasurekar@nvidia.com> * Add SHM size 16gb to reranking and VLM NIM in compose (#405) * Fixed the query to be derived from messages if query is not explicitly given in argument and adding messages list in logs (#404) * Added url for nvidia/llama-nemotron-rerank-1b-v2 model for cloud endpoint (#412) Signed-off-by: Swapnil Masurekar <smasurekar@nvidia.com> * fix start proble failureThreshold to 750 (#411) * Update: Remove Vss on AIDP notebook (#409) * Update: Remove Vss Signed-off-by: Minh Nguyen <minhngu@nvidia.com> * Remove video processing (VSS) from kafka consumer Video handler, video analyzer service, and all VSS-related configuration have been removed to simplify the event ingestion pipeline to document-only processing. Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Made-with: Cursor Signed-off-by: Minh Nguyen <minhngu@nvidia.com> * fix: add USERID Signed-off-by: Minh Nguyen <minhngu@nvidia.com> * fix: update document Signed-off-by: Minh Nguyen <minhngu@nvidia.com> --------- Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Co-authored-by: Minh Nguyen <minhngu@nvidia.com> * Kheiss/chunking topic (#417) * confirming presence of switcher text in conf.py file * Added chunking information * CI - Updated Nemotron endpoints for embedding, page-elements, graphic-elements, table-structure (#410) Signed-off-by: Swapnil Masurekar <smasurekar@nvidia.com> Co-authored-by: Shubhadeep Das <149712532+shubhadeepd@users.noreply.github.com> * Add notebook showcasing langchain connector for Nvidia RAG Retrieval (#419) * notebook: Add notebook showcasing langchain retriever connector * Update langchain connector version to 1.2.0 * Fix broken link in notebook doc * Revert "Added url for nvidia/llama-nemotron-rerank-1b-v2 model for cloud endp…" and add ranker endpoint in nvdev (#421) This reverts commit 1a11733169fb17e80537f7ded879472a1ada4d8c. Signed-off-by: Swapnil Masurekar <smasurekar@nvidia.com> * Nemoretriever OCR version 1.2.0 -> 1.2.1 in Helm (#422) * Fix name for rag langchain connector (#423) * Print reasoning tokens if DEBUG logging is enabled (#424) * Print reasoning tokens if DEBUG logging is enabled (#378) * Print reasoning tokens if VERBOSE logging is enabled Signed-off-by: Niyati Singal <nsingal@nvidia.com> * fix(observability): use hardcoded GenAI attribute keys instead of deprecated SpanAttributes (#377) Signed-off-by: Swapnil Masurekar <smasurekar@nvidia.com> --------- Signed-off-by: Niyati Singal <nsingal@nvidia.com> Signed-off-by: Swapnil Masurekar <smasurekar@nvidia.com> Co-authored-by: Swapnil Masurekar <smasurekar@nvidia.com> * Print reasoning tokens if DEBUG logging is enabled Signed-off-by: Niyati Singal <nsingal@nvidia.com> --------- Signed-off-by: Niyati Singal <nsingal@nvidia.com> Signed-off-by: Swapnil Masurekar <smasurekar@nvidia.com> Co-authored-by: Swapnil Masurekar <smasurekar@nvidia.com> * CI: Update helm packaging and selective publish support (#425) * Add rag-blueprint agent skill with CLAUDE.md and project config (#407) * Add rag-blueprint agent skill with CLAUDE.md and project config * Update query-and-conversation.md * Update notebooks.md Updated with NVIDIA style guide recommendations --------- Co-authored-by: Kurt Heiss <kheiss@nvidia.com> * doc: Fix broken links in notebooks and docs (#426) * Kheiss/prd additions (#427) * confirming presence of switcher text in conf.py file * Update documentation to reflect name change to NeMo Retriever Library * Update api-rag.md removed GitHub markers * Update change-model.md Removed GitHub markers * Update deploy-helm.md removed GitHub markers * Update deploy-helm.md Remove GitHub markers * Update deploy-helm.md Remove GitHub markers * Update mig-deployment.md removed GitHub markers * Update deploy-helm.md removed extra space * Launchable Updates for release 2.5 (#428) (#429) Co-authored-by: rkharwar-nv <rkharwar@nvidia.com> * Kheiss/cont ingest (#431) * confirming presence of switcher text in conf.py file * Continuous ingestion topic * updated index for continuous ingestion * updated reademe for continuous ingestion * Update continuous-ingestion-object-storage.md added RAG Blueprint * Update continuous-ingestion-object-storage.md Converted first sentence into 2 sentences * Update index.md * Apply suggestion from @nkmcalli Co-authored-by: nkmcalli <nmcallister@nvidia.com> * Update continuous-ingestion-object-storage.md * Update continuous-ingestion-object-storage.md * Update continuous-ingestion-object-storage.md --------- Co-authored-by: nkmcalli <nmcallister@nvidia.com> * Nemotron 3 super deployment guide and migration guide (#430) * Nemotron 3 super deployment guide and migration guide * Organize gpu requirement and heading for nemotron3 super * Add instruction for updating values.yaml and refractor doc * Add cloud endpoint url in env file * Instruction to export llm max token in docker flow * remove unrequired llm api key from doc * Seprate yaml for nemotron 3 deployment * Remove unnecessary information for local hosted * Simplify docker deployment logs * Remove cuda device from rtx 6000 pro * Add prompt customization instruction in nemotron3 helm section * Instruction for prompt customization * Remove heading for rtx 6000 pro * docs: move NIM_MAX_MODEL_LEN and LLM_MAX_TOKENS to general self-hosted deployment steps (#432) * Update: Remove vss and update Minio access console (#433) * VLM embed doc fix (#435) Signed-off-by: smasurekar <smasurekar@nvidia.com> * docs: add RAG accuracy benchmarks documentation (#434) * docs: add RAG accuracy benchmarks documentation * docs: Fix broken links and format in accuracy benchmark doc * Update accuracy-benchmarks.md * Update accuracy-benchmarks.md * Update accuracy-benchmarks.md * Update accuracy-benchmarks.md * Update accuracy-benchmarks.md * Update accuracy-benchmarks.md implmented changes as instructed by Sumit in Slack thread: https://nvidia.slack.com/archives/C09HAQRT1UY/p1773470561423909 --------- Co-authored-by: Kurt Heiss <kheiss@nvidia.com> * Update to GA artifact path (#437) * Update changelog to include new additions * Update containers to GA version * adding missing accuracy benchmark documentation (#438) * adding missing accuracy benchmark documentation * Update docs/evaluate.md Co-authored-by: nkmcalli <nmcallister@nvidia.com> --------- Co-authored-by: nkmcalli <nmcallister@nvidia.com> * fix(unit): avoid real Milvus in delete_documents tests for CI (#440) Signed-off-by: Swapnil Masurekar <smasurekar@nvidia.com> * Update release date for v2.5.0 * fix: validate file paths in MCP upload/update tools to prevent path traversal (CWE-22) The tool_upload_documents and tool_update_documents MCP tools accepted arbitrary file paths from MCP clients without validation. An attacker controlling the MCP client (or an LLM agent making tool calls) could supply paths like /etc/shadow, /proc/self/environ, or ../../sensitive.yaml to read arbitrary files from the server filesystem and exfiltrate them by uploading to the ingestor. Add _validate_file_path() helper that resolves paths via os.path.realpath() (following symlinks) and verifies they reside within the allowed upload directory (MCP_UPLOAD_DIR env var, defaults to cwd). Raises ValueError for paths outside the sandbox. Both tool_upload_documents and tool_update_documents now call this validator before reading any file. Signed-off-by: Sebastion <sebastiondev@users.noreply.github.com> * Kheiss/versioning modifiers (#462) * Updated Vidore Dataset to Vidore V3 Dataset (#443) * Kheiss/rm early access1 (#445) * remove early access from title * remove early access from title * Update documentation per broken link reporting for Brev and support matrix (#458) * updated versioning method for RAG documentation * Add GCNV data ingestor Helm chart example (#475) Package the GCNV data ingestor deployment into a reusable Helm chart with PVC, service, and namespace templates plus installation guidance for Trident-backed storage. Made-with: Cursor Signed-off-by: Raj Sahoo <raj.sahoo@netapp.com> * docs(perf): add RAG performance measurement methodology (#490) * docs(perf): add RAG performance measurement methodology Add performance benchmarking documentation covering TTFT and ITL metrics across four datasets (KG-RAG, RagBattlePacket, HotPotQA, BO767) plus synthetic workloads, comparing LLM-49B and VLM nano configurations with reasoning on/off. Signed-off-by: Truong Nguyen <tgnguyen@nvidia.com> * Update docs/perf-benchmarks.md Co-authored-by: Kurt Heiss <kheiss@nvidia.com> * Update docs/perf-benchmarks.md Co-authored-by: Kurt Heiss <kheiss@nvidia.com> * Update docs/perf-benchmarks.md Co-authored-by: Kurt Heiss <kheiss@nvidia.com> * Update docs/perf-benchmarks.md Co-authored-by: Kurt Heiss <kheiss@nvidia.com> * Update docs/perf-benchmarks.md Co-authored-by: Kurt Heiss <kheiss@nvidia.com> * Update docs/perf-benchmarks.md Co-authored-by: Kurt Heiss <kheiss@nvidia.com> * Update docs/perf-benchmarks.md Co-authored-by: Kurt Heiss <kheiss@nvidia.com> * Update docs/perf-benchmarks.md Co-authored-by: Kurt Heiss <kheiss@nvidia.com> * Update docs/perf-benchmarks.md Co-authored-by: Kurt Heiss <kheiss@nvidia.com> * Update docs/perf-benchmarks.md Co-authored-by: Kurt Heiss <kheiss@nvidia.com> --------- Signed-off-by: Truong Nguyen <tgnguyen@nvidia.com> Co-authored-by: Kurt Heiss <kheiss@nvidia.com> * ci: add Run Branch Script dispatcher workflow (#548) Adds a generic workflow_dispatch workflow that checks out any branch and runs a specified script on the self-hosted GPU runner. Enables iterating on NV-BASE eval CI (feat/nvbase-ci-smoke) without merging workflow changes to main for every iteration. Co-authored-by: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com> * ci: make runner configurable in Run Branch Script dispatcher (#550) Adds a runner input (choice) so the target runner can be changed without another PR. Switches default to sonarqube-workflows-bp-sre (internally hosted, no GPU) since arc-runners-org-nvidia-ai-bp-2-gpu is cloud-hosted and cannot reach urm.nvidia.com (NVIDIA internal Artifactory) needed to install nv-base. Co-authored-by: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com> * ci: add blueprints-skills-eval-runner as default runner New dedicated internal VM for RAG skill eval CI. Added as default runner option in the dispatcher workflow. sonarqube-workflows-bp-sre kept as fallback. Co-Authored-By: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com> * ci: add rag-eval runner option to Run Branch Script dispatcher (#558) Adds 'rag-eval' to the runner input's choice list so workflow_dispatch can target the dedicated rag-skill-validator self-hosted runner (labels: self-hosted, Linux, X64, rag-eval, brev, rag). This mirrors the VSS skill-eval pattern where a dedicated label routes jobs to a Brev CPU coordinator VM. Used by the upcoming ci/run_skill_eval.sh script on feat/skill-eval. Signed-off-by: richa <ricsingh@nvidia.com> Co-authored-by: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com> * RAG v2.5.1 hotfix: release prep follow-ups (BCS-445) (#505) * Feat: Update Nemotron-Omni model config * Update: Update tag v2.5.1 document * chore: regen uv.lock, sweep docs scripts, sync helm prompt (BCS-445) Signed-off-by: Phuc Nguyen <pdnguyen@nvidia.com> * fix: bump langchain-core/orjson/pillow for nSpect CVEs Signed-off-by: Phuc Nguyen <pdnguyen@nvidia.com> * Promote nemotron-omni VLM to GA image and switch VLM/image-extract defaults to opt-in - Bump VLM image: nvcr.io/nvstaging/nim/nemotron-3-nano-30b-a3b-omni-reasoning:1.7.0-variant-rc2-48948519 -> nvcr.io/nim/nvidia/nemotron-3-nano-omni-30b-a3b-reasoning:1.7.0-variant (RC removed; GA word order moves "omni" before 30b-a3b in both image path and model name) - Rename model identifier across compose, helm values, src docstrings, notebooks and docs - Default ENABLE_VLM_INFERENCE -> False (helm + compose) - Default APP_NVINGEST_EXTRACTIMAGES -> False (helm + compose) - VLM_TO_LLM_FALLBACK confirmed False (no change) - Update release notes / vlm.md to drop "RC2 not pullable" caveat now that GA is live Signed-off-by: Phuc Nguyen <pdnguyen@nvidia.com> * revert: keep text-only embedder as default for v2.5.1 Per PR review feedback (#505): the default embedding model in 2.5.0 was nvidia/llama-nemotron-embed-1b-v2 (text-only). To preserve backward compatibility for existing 2.5.0 collections in the v2.5.1 release, restore the text-only embedder as the default everywhere and make the VLM embedder (llama-nemotron-embed-vl-1b-v2) opt-in. Embedder defaults reverted to 2.5.0: - deploy/compose/.env, nvdev.env, variables.env: text endpoint + model - docker-compose-rag-server.yaml + ingestor-server.yaml: text active, VLM commented out - deploy/workbench/compose.yaml: text endpoint + model - nims.yaml: text profile = ["", "rag", "ingest", "text-embed", "vlm-generation"] (default), VLM profile = ["vlm-embed", "vlm-ingest"] (opt-in) - helm values.yaml: APP_EMBEDDINGS_* + EMBEDDING_NIM_* point to text; nvidia-nim-llama-32-nv-embedqa-1b-v2.enabled=true, nvidia-nim-llama-nemotron-embed-vl-1b-v2.enabled=false - endpoints.md: docs reflect text-only default Other follow-ups in this PR: - prompt.yaml (helm + src): restore "/no_think" system directive on image_captioning_prompt that was removed during VLM swap - docs/release-notes.md: rewrite v2.5.1 highlights to describe Nemotron- Omni as opt-in (ENABLE_VLM_INFERENCE=False default), drop the breaking- change list that no longer applies (no vector-dim change, no profile remap, no /no_think removal) VLM (Nemotron-Omni) opt-in path remains intact; image, profile aliases, and helm sub-chart definitions are unchanged - only the defaults flip. Signed-off-by: Phuc Nguyen <pdnguyen@nvidia.com> * chore: address PR review polish — nim-vlm default false + trim release notes - helm values.yaml: nim-vlm.enabled = false (VLM is opt-in) - release-notes.md: drop opening "New opt-in VLM" highlight and the closing "Default embedder unchanged" note (covered in lead paragraph) Signed-off-by: Phuc Nguyen <pdnguyen@nvidia.com> * revert: keep ingestor-server + rag-frontend at 2.5.0 (reuse from previous release) Only rag-server is bumped to 2.5.1 in this hotfix (per BCS-445 scope). Per Shubhadeep's PR review: ingestor-server and rag-frontend containers are unchanged in v2.5.1 and should reuse the existing 2.5.0 images on NGC. Reverts the inadvertent ${TAG:-2.5.1} default in: - deploy/compose/docker-compose-ingestor-server.yaml (ingestor-server) - deploy/compose/docker-compose-rag-server.yaml (rag-frontend) - deploy/workbench/compose.yaml (ingestor-server + rag-frontend) Helm values.yaml already correct (ingestor-server + rag-frontend tags at "2.5.0"). --------- Signed-off-by: Phuc Nguyen <pdnguyen@nvidia.com> Co-authored-by: An Nguyen <anngu@nvidia.com> * feat(ci): enable copy-pr-bot for automated PR mirroring Activates copy-pr-bot on NVIDIA-AI-Blueprints/rag. All NVIDIA org members with write access are auto-trusted — no individual names needed. Requires signed commits for trusted-change classification. Must be on the default branch (main) per copy-pr-bot requirements. The skill eval workflows (skills-eval.yml) live on develop where PRs are based. Co-Authored-By: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com> Signed-off-by: Vidushi Gupta <vidushig@nvidia.com> * chore(ci): add vidushig-nv and richa-nvidia as additional_trustees (#604) Needed because author_association is CONTRIBUTOR not MEMBER for users who are not direct members of NVIDIA-AI-Blueprints org. This ensures their signed commits auto-mirror without needing /ok to test. Signed-off-by: Vidushi Gupta <vidushig@nvidia.com> Co-authored-by: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com> * ci: add NVSkills CI request workflow (signing pipeline) (#611) Adds .github/workflows/request-nvskills-ci.yml — the source-repo handler for the NVSkills CI / NVCARPS signing pipeline. Triggered by maintainer or admin commenting /nvskills-ci on a PR that changes skills/, and by the signature-push back from nv-nvskill-ci[bot]. Companion to NVBug 6209970 (App install + onboarding + dispatch token) and PR #603 (canonical skills/ layout migration), ahead of the 2026-05-27 Computex cutover for the NVIDIA Verified Skills catalog. Signed-off-by: Pranjal Doshi <pranjald@nvidia.com> * ci: Fix nv skills bot name (#626) * ci(cve): mirror Phase-3 PR-creation workflow to main This brings the GitHub Actions side of the nightly CVE Phase-3 flow onto main, which is where workflow_dispatch needs the file to live for the manual "Create PR" button on Issue #617 to actually trigger. Files mirrored from develop (merged via #636): - .github/workflows/cve-create-pr.yml — workflow_dispatch entry point, runs on blueprints-skills-eval-runner; reads the patch from the marker comment on Issue #617, applies to a fresh develop checkout, calls cve-create-pr.sh to materialize branch + PR - .github/scripts/cve-create-pr.sh — Git Data API dance: blobs → tree → commit → ref → PR. Idempotent on same-day re-clicks The cve-post script (ci/post-cve-report.sh) stays on develop only — GitLab clones it from there during cve-scan and bundles it into the artifact. After this lands on main, the maintainer-clicked "Create PR" button on Issue #617 becomes functional. No nightly automation changes. Signed-off-by: richa <ricsingh@nvidia.com> * ci(cve-pr): mirror compare-URL fallback workflow to main Companion to PR #642 (target: develop) which adds the same changes. This PR brings the GHA workflow + script onto main, which is where GitHub Actions loads the workflow YAML from for workflow_dispatch. Without this, the new "Notify Issue" step and the script's compare-URL emission never take effect — the running workflow is whatever's currently on main. Why two PRs (develop + main): - ci/post-cve-report.sh runs from develop (GitLab cve-post clones origin/develop). Only on develop in PR #642 - .github/workflows/cve-create-pr.yml is loaded by GitHub from the default branch (main) when workflow_dispatch fires. Must be on main - .github/scripts/cve-create-pr.sh is executed by the workflow after it checks out develop, so develop's copy is what runs. But main is kept in sync to avoid drift Same content as PR #642 for these two files. After both merge, the maintainer-clicked Create-PR flow works end-to-end under the restrictive org policy that blocks gh pr create: click 🚀 → workflow runs ~2 min → Issue notification with compare URL → click → GitHub compare page (pre-filled) → click Create PR. See PR #642 description for the full flow + verification notes. Signed-off-by: richa <ricsingh@nvidia.com> * ci: accept maintain role for nvskills requests (#647) Signed-off-by: Pranjal Doshi <pranjald@nvidia.com> * ci(cve): edit-in-place compare-URL comment via HTML marker Replace the append-only "Notify Issue with compare URL" step with an edit-in-place rolling comment, tagged by . Mirrors the pattern already used in ci/post-cve-report.sh for the  marker comment. Before: each successful workflow_dispatch click left a new comment on Issue #617 (one per branch per click). Same-day re-runs duplicated the comment; same-day re-runs with a force-updated branch also produced stale commit SHAs in earlier comments (the SHA no longer matched the branch tip). After 30 nightlies, 30+ comments accumulated. After: a single rolling comment is updated on every run. Same-day re-runs silently overwrite the previous compare-URL line; multi-day runs show only the latest branch (older branches are still in Git, recoverable via `git branch -r | grep cve-fix/nightly-`). Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> Signed-off-by: richa <ricsingh@nvidia.com> * [v2.6.0] Sync release-v2.6.0 to main (#657) * chore: prepare release-v2.6.0 sync to main Signed-off-by: Shubhadeep Das <shubhadeepd@nvidia.com> * chore: keep release image paths staged Signed-off-by: Shubhadeep Das <shubhadeepd@nvidia.com> * Helm: expose podAnnotations on all NIMService templates (#658) Plumb a per-NIM podAnnotations field from values.yaml through to NIMService.spec.podAnnotations so users can attach pod-level annotations to NIM workloads. Default is {} (omits the field), so existing deployments render identically. Primary motivator is Runai fractional GPU saving-mode, which requires both gpu-fraction-style annotations on the pod AND fractional GPU resources, e.g.: nimOperator: nim-llm: podAnnotations: gpu-fraction: "0.25" gpu-fraction-num-devices: "1" resources: limits: { runai.com/gpu: 1 } requests: { runai.com/gpu: 1 } Templates touched: llm-nim, embedding-nim, reranking-nim, vlm-nim, vlm-captioning-nim, vlm-embed-nim, vlm-reranker-nim. Each gains the podAnnotations: {} default and a usage comment in values.yaml. (cherry picked from commit ab4cddf430a7d8c8e3e05c2e64685ad5aeeaf1e3) Signed-off-by: Nikhil Kulkarni <nikkulkarni@nvidia.com> Co-authored-by: Nikhil Kulkarni <nikkulkarni@nvidia.com> (cherry picked from commit b1ea5e813909e81bad2d96a54c106f34cff658a5) Signed-off-by: Shubhadeep Das <shubhadeepd@nvidia.com> * fix: move vlm reranker host port (#656) Signed-off-by: Shubhadeep Das <shubhadeepd@nvidia.com> (cherry picked from commit 51d5caf79fcd17def62aa40e751ee4955323baa3) Signed-off-by: Shubhadeep Das <shubhadeepd@nvidia.com> * [codex] Refresh v2.6 documentation support guidance (#659) * docs: refresh v2.6 support guidance Signed-off-by: Shubhadeep Das <shubhadeepd@nvidia.com> * docs: tighten reasoning and mig guidance Signed-off-by: Shubhadeep Das <shubhadeepd@nvidia.com> --------- Signed-off-by: Shubhadeep Das <shubhadeepd@nvidia.com> (cherry picked from commit 1075cb3f055a3611941926a40381674b9cabbfe0) Signed-off-by: Shubhadeep Das <shubhadeepd@nvidia.com> * chore: update Vite lockfile to 6.4.2 (#660) Signed-off-by: Shubhadeep Das <shubhadeepd@nvidia.com> (cherry picked from commit e5602db19f4f1977683296c692fdf87b74e7e3ef) Signed-off-by: Shubhadeep Das <shubhadeepd@nvidia.com> * chore: update blueprint container registry paths (#661) * chore: update blueprint container registry paths Signed-off-by: Shubhadeep Das <shubhadeepd@nvidia.com> * ci: tag publish images for staging registry Signed-off-by: Shubhadeep Das <shubhadeepd@nvidia.com> --------- Signed-off-by: Shubhadeep Das <shubhadeepd@nvidia.com> (cherry picked from commit 8b6f492ea2583c5c8e87d71d4e3621bcf50eeea9) Signed-off-by: Shubhadeep Das <shubhadeepd@nvidia.com> * ci: enable coderabbit reviews for draft prs Signed-off-by: Shubhadeep Das <shubhadeepd@nvidia.com> --------- Signed-off-by: Shubhadeep Das <shubhadeepd@nvidia.com> Signed-off-by: Nikhil Kulkarni <nikkulkarni@nvidia.com> Co-authored-by: Nikhil Kulkarni <nikkulkarni@nvidia.com> * docs: fix Sphinx build, version switcher, and style for v2.6.0 (#665) Resolve broken xrefs and lexers, add version switcher manifest hook, update versions1.json for 2.6.0-2.3.0, and apply DORI style guide fixes. (cherry picked from commit f66cdec10d5e5c2d4a77d4dde4a4b4dea37797fa) Signed-off-by: Kurt Heiss <kheiss@nvidia.com> Co-authored-by: Kurt Heiss <kheiss@nvidia.com> * deploy RAG stack directly via brev exec, not Harbor; add TaskOutput t… (#668) * deploy RAG stack directly via brev exec, not Harbor; add TaskOutput timeout; reduce GPU trial multiplier * fix: address CodeRabbit comments — env field clarity, TaskOutput timeout behavior, robust NIM check Signed-off-by: Vidushi Gupta <vidushig@nvidia.com> * fix(eval): discover NIM model dynamically instead of hardcoding model name Signed-off-by: Vidushi Gupta <vidushig@nvidia.com> --------- Signed-off-by: Vidushi Gupta <vidushig@nvidia.com> * chore(release-v2.6.0): sync latest develop changes (#666) (#667) * ci: minimal Issue body — status + links, full report in artifact The previous design embedded the skill's full _summary.md (including Phase 1 fetch details, Phase 6 reviewer verdicts, Infrastructure Map, Active Versions Checked, NVBugs Audit Trail, etc.) into the GitHub Issue body. ~100 lines of skill-internal jargon for a daily glance. New design: 10-line body — status badge (✅ / ⚠️ / ℹ️ derived from the **Counts:** line in _summary.md), date, scanned commit, GitLab pipeline link, and Full report (artifact) link. Per-CVE detail, reviewer verdicts, manifest diff, validation results live in the GitLab artifact — one click from the body. Also reads SCAN_JOB_URL from the dotenv emitted by cve-scan (see companion change on GitLab ci/nightly-cve), so the artifact link points at the scan job (where the artifact lives), not cve-post's own job URL. * ci: auto-reopen tracker Issue if closed before refreshing body If anyone accidentally closes #617, the next nightly run re-opens it so the rolling tracker can never be orphaned. gh issue reopen is idempotent — no-op when the Issue is already open, prints "Issue is already open" and exits 0. Errors are swallowed so this is purely best-effort defensive. * feat(skills): migrate to canonical skills/ path per publishing guide Move skill content from skill-source/.agents/skills/{rag-blueprint,rag-eval, rag-perf}/ to skills/{rag-blueprint,rag-eval,rag-perf}/ at the repo root, matching the agentskills.io canonical layout (Step 4 of the Skills Publishing Onboarding Guide). The skill-source/ directory is removed entirely; audit confirmed no internal or external tool hard-codes the old path. Also: - Move skill-source/README.md -> skills/README.md - Move skill-source/validate_skill_api_versions.py -> scripts/validate_skill_versions.py - Rename tests/unit/test_skill_source/ -> tests/unit/test_skills/ - Add BENCHMARK.md to each of the three skills (Harbor-based methodology) - Add skills/PUBLISHING_COMPLIANCE.md as the single source of truth for the remaining external coordination items (NVCARPS signing onboarding with a 2026-05-27 deadline for the Computex cutover, catalog PR draft, OSRB checklist, NV-BASE runner install) - Update all internal references across CLAUDE.md, AGENTS.md, README.md, ci/run_skill_eval.sh, skill-eval/, .openclaw/, .github/skill-eval/AGENTS.md, .github/workflows/skills-eval.yml, docs/release-notes.md Companion change to be opened against NVIDIA/skills:components.d/rag-blueprint.yml to update the path field from skill-source/.agents/skills/ to skills/. * agentic-bugfix: NVBug 6229456 * Fix: Reduce Otel batch size (#628) * ci: include Triage table with absolute artifact URLs Re-add the Triage section to the Issue body so maintainers see the at-a-glance per-package breakdown without leaving GitHub. Rewrites the per-CVE relative markdown links (which break when clicked from GitHub) to absolute GitLab artifact URLs — clicking goes directly to the file in GitLab's artifact viewer. Extracts from "## Triage" through the next "## " heading, so the Counts + Result lines come along but the deeper sections (Phase 1 fetch details, Phase 6 reviewer verdicts, Infrastructure Map, etc.) stay in the artifact. Verified locally against both yesterday's 6-CVE artifact and a synthetic 0-CVE summary. * Helm MIG (H100): promote embedding-VLM to a full GPU (#633) * Helm MIG (H100): promote embedding-VLM to a full GPU Reshape the H100 MIG layout so the embedding-VLM NIM gets a dedicated full H100 instead of a 3g.40gb slice, improving vision-tower throughput. GPU 0,1 — full devices → nim-llm (vLLM tp=2) GPU 2 — 1x 2g.20gb + 3x 1g.10gb + 1x 1g.20gb → OCR + (graphic, page, table) + rerank GPU 3 — full device → embedding-VLM (NEW) Rerank keeps its 1g.20gb slice (now on GPU 2). OCR drops from 3g.40gb to 2g.20gb to make room for rerank's slice on GPU 2 (no spare slots remain). MIG profile renamed to custom-h100-llm2full-embed1full-1x2g20-3x1g10-1x1g20 and docs (mig-deployment.md) updated to match the new layout and verification output. * Helm MIG (H100): expand to 5-GPU layout to restore OCR slice Adding a 5th H100 dedicated to the reranker lets the embedding-VLM keep its full GPU (GPU 3) while restoring OCR to its original 3g.40gb slice on GPU 2. The previous 4-GPU layout had forced OCR down to 2g.20gb to fit rerank's 1g.20gb on GPU 2 (H100 MIG memory-slice budget of 8 was exhausted when packing OCR + 3 small NIMs + rerank on one GPU). GPU 0,1 — full devices -> nim-llm (vLLM tp=2) GPU 2 — 1x 3g.40gb + 4x 1g.10gb -> OCR + graphic/page/table (+ spare) GPU 3 — full device -> embedding-VLM GPU 4 — 1x 3g.40gb + 2x 1g.20gb -> rerank (+ spare 3g.40gb and 1g.20gb) MIG profile renamed to custom-h100-5gpu-llm2full-embed1full and docs/mig-deployment.md updated (layout description, kubectl labels, kubectl-view-allocations, and nvidia-smi -L verification output). --------- * ci(cve): add Review & Create-PR flow to nightly tracker Issue Phase 3 of the nightly CVE pipeline. When the agentic-cve-fix skill produces workspace edits (non-empty cve-fix.patch), this flow lets a maintainer review the proposed diff inline on Issue #617, then click a single button to materialize the fix as a PR on develop. GitLab side (cve-post / ci/post-cve-report.sh): - Detect cve-fix.patch in the artifact root - Edit-or-create a single hidden-marker comment () on the tracker Issue with the patch inline in a <details> block - Append "Review & Create PR" section to the body with: * Review link → #issuecomment-<id> anchor on the same Issue * Create PR link → workflow_dispatch URL of cve-create-pr.yml - 60 KB inline threshold; fall back to a GitLab artifact link when over GitHub side (.github/workflows/cve-create-pr.yml + script): - workflow_dispatch on blueprints-skills-eval-runner (verified via probe to reach github.com and gitlab-master.nvidia.com) - Read patch from the marker comment via gh api - git apply --3way on a fresh develop checkout - Build signed commit through Git Data API (gh api git/blobs, trees, commits, refs) — commits are auto-signed by github-actions[bot] - Idempotent: existing PR for cve-fix/nightly-<date> short-circuits No new tokens, secrets, or branches. GitLab GITHUB_PAT stays at Issues:write only; GitHub workflow uses the default GITHUB_TOKEN. Validated locally with shellcheck and 5 mock-gh integration tests (first run, second run / edit, empty patch, full PR creation, re-click idempotency). * ci(cve): install gh CLI on runner before Git Data API calls The blueprints-skills-eval-runner does not ship with the gh CLI. Probe (run 26561466990) caught this — the workflow would have failed at the first gh api call with command-not-found. Install pattern: download pinned gh 2.62.0 tarball to RUNNER_TEMP/bin and add to GITHUB_PATH. Subsequent steps in the job see gh on PATH. Idempotent: if gh is later installed system-wide, command -v short- circuits the download. Workflow-scoped temp dir means no runner state pollution between jobs. Validated by probe v2 (run 26561674766): gh 2.62.0 installs and runs on the real runner. * skill maintainence: block PRs where changed skill has no eval/ directory * ci(cve-post): rewrite ./FILENAME.md Triage links too Skill commit f5bb788d (jarvis/ai_rules main, 2026-05-27) changed the Triage table link format from `(FILENAME.md)` to `(./FILENAME.md)`. The earlier link-rewrite regex `[^()/]+\.md` excluded `/` so it never matched the new format, leaving 8 broken relative links in today's Issue #617 body (pipeline 52917159 output). Update the regex to accept an optional `./` prefix: s|$(\.?/?)([^()/]+\.md)$|(${URL_PREFIX}/\2)|g Group 2 captures just the bare filename, so the rewritten URL doesn't inherit the `./` segment. Mirror the same shape in the SCAN_JOB-absent fallback branch (which strips broken links rather than rewriting). Verified locally with three test cases: [a](a.md) -> rewritten [b](./b.md) -> rewritten [c](https://x.com/y) -> untouched (absolute URLs pass through) Branch dedicated to this fix so we can test against the live Issue artifact before merging. * ci(cve-pr): switch from gh pr create to compare-URL fallback NVIDIA-AI-Blueprints org policy blocks GHA workflows from calling gh pr create via GITHUB_TOKEN ("createPullRequest" forbidden). The permissions: block has no effect — the org-level setting overrides repo-level. Switching to a service-account PAT is multi-day work. Workaround: the branch + signed commit are already pushed by the Git Data API path (which works under restrictive permissions). Build a https://github.com/.../compare/develop...BRANCH?expand=1 URL with title and body pre-filled via query params. The maintainer clicks once, GitHub opens its native PR-creation page with everything populated, they click "Create pull request" — PR exists. Script changes (.github/scripts/cve-create-pr.sh): - Replace gh pr create block with compare-URL emission - URL-encode title + body via jq -sRr @uri (handles spaces, newlines, backticks, &, #, etc.) - 6 KB length cap with graceful body-drop fallback - Write summary banner to GITHUB_STEP_SUMMARY with the URL as a styled clickable link - Export compare_url, branch_name, commit_sha to GITHUB_OUTPUT - Empty-diff abort (NEW): if git apply leaves no workspace changes (develop already has the fix), exit 0 with a clear banner instead of failing the workflow - Existing-PR branch now also writes a "View existing PR" banner to the summary so re-triggers show the URL even after the previous run's page closed Workflow changes (.github/workflows/cve-create-pr.yml): - Add issues:write to permissions block — needed for the new step - Add "Notify Issue" step after the script step. Posts a comment on the tracker Issue with the compare URL so anyone subscribed gets a notification (the workflow Summary alone is easy to miss) - Step uses id: create on the script run + reads its outputs - Apostrophe-free heredoc body for bash 3.2 portability (apostrophes inside $(cat <<EOF) misparse on bash 3.2) Trade-off accepted: one extra human click per CVE fix (compare page's "Create pull request" button). The PR is opened under the maintainer's identity, which is actually a feature — clear audit trail of human approval. Validated with: shellcheck clean, YAML parse clean, bash 3.2 portability tested, hostile code review (caught the apostrophe issue), runner reachability tested (probe v2 + smoke tests). * agentic-bugfix: NVBug 6229403 | Brev launchable fix (#630) * agentic-bugfix: NVBug 6229403 * Remove depricated changes from launchable notebook * Added min. disk space for docker as 150gb in brev launchable --------- * Attach NVSkills validation signatures * Helm MIG (RTX 6000): drop nim-llm gpus product selector (#648) Remove the `gpus: [- product: "rtx6000_blackwell_sv"]` selector from nimOperator.nim-llm.model. The remaining engine/precision/tensorParallelism (vllm / fp8 / tp2) already matches profile a89dbe9e… on its own, and keeping the product selector was over-constraining model selection without adding correctness. * ci(cve): mint GitHub App installation token in cve-post script When NV_RAG_CVE_BOT_APP_ID, NV_RAG_CVE_BOT_INSTALLATION_ID, and NV_RAG_CVE_BOT_PRIVATE_KEY are set, post-cve-report.sh now mints a short-lived (~9 min) installation token via RS256 JWT signing and uses it in place of GITHUB_PAT. This flips the actor on Issue #617 body edits and marker-comment writes from a personal user identity to the nv-rag-cve-bot App, removing personal-name attribution and enabling GitHub notifications to the maintainer. The implementation is additive: if any of the three App env vars are unset (local dev, pre-migration smoke runs), the script falls back to the existing GITHUB_PAT path. The rest of the script is unchanged — it consumes GITHUB_PAT uniformly, regardless of which auth mode produced the token. The companion GitLab .gitlab-ci.yml change (apk add python3 py3-cryptography in cve-post + cve-smoke-test) is already on ci/nightly-cve as commit 8edfa6e. Verified end-to-end against the live App (3902749) installed on this repo: JWT mint succeeds, token returns permissions=contents:write, issues:write, metadata:read, pull_requests:write scoped to this repo only; real write test on Issue #617 attributed to user.login='nv-rag-cve-bot[bot]', user.type='Bot'. * ci(cve): edit-in-place compare-URL comment via HTML marker Replace the append-only "Notify Issue with compare URL" step with an edit-in-place rolling comment, tagged by . Mirrors the pattern already used in ci/post-cve-report.sh for the  marker comment. Before: each successful workflow_dispatch click left a new comment on Issue #617 (one per branch per click). Same-day re-runs duplicated the comment; same-day re-runs with a force-updated branch also produced stale commit SHAs in earlier comments (the SHA no longer matched the branch tip). After 30 nightlies, 30+ comments accumulated. After: a single rolling comment is updated on every run. Same-day re-runs silently overwrite the previous compare-URL line; multi-day runs show only the latest branch (older branches are still in Git, recoverable via `git branch -r | grep cve-fix/nightly-`). * Helm: expose podAnnotations on all NIMService templates (#645) Plumb a per-NIM podAnnotations field from values.yaml through to NIMService.spec.podAnnotations so users can attach pod-level annotations to NIM workloads. Default is {} (omits the field), so existing deployments render identically. Primary motivator is Runai fractional GPU saving-mode, which requires both gpu-fraction-style annotations on the pod AND fractional GPU resources, e.g.: nimOperator: nim-llm: podAnnotations: gpu-fraction: "0.25" gpu-fraction-num-devices: "1" resources: limits: { runai.com/gpu: 1 } requests: { runai.com/gpu: 1 } Templates touched: llm-nim, embedding-nim, reranking-nim, vlm-nim, vlm-captioning-nim, vlm-embed-nim, vlm-reranker-nim. Each gains the podAnnotations: {} default and a usage comment in values.yaml. * ci: add ci/deploy/compose to pre-checkout cleanup for root-owned volumes --------- (cherry picked from commit c47aa2bbde307c693b5fa2e8b8810d48545d8f72) Signed-off-by: richa <ricsingh@nvidia.com> Signed-off-by: shubhadeepd <shubhadeepd@local> Signed-off-by: Nikhil Kulkarni <nikkulkarni@nvidia.com> Signed-off-by: Richa Singh <ricsingh@nvidia.com> Signed-off-by: agentic-bug-fix <agentic-bug-fix@local> Signed-off-by: smasurekar <smasurekar@nvidia.com> Signed-off-by: nvskills-svc-account <svc-nvskills-signing@nvidia.com> Signed-off-by: Vidushi Gupta <vidushig@nvidia.com> Signed-off-by: Shubhadeep Das <shubhadeepd@nvidia.com> Co-authored-by: richa <ricsingh@nvidia.com> Co-authored-by: Claude Opus 4.7 <noreply@anthropic.com> Co-authored-by: Pranjal Doshi <pranjald@nvidia.com> Co-authored-by: shubhadeepd <shubhadeepd@local> Co-authored-by: anngu-2xx3 <anngu@nvidia.com> Co-authored-by: nv-pranjald <150428320+nv-pranjald@users.noreply.github.com> Co-authored-by: Nikhil Kulkarni <nikkulkarni@nvidia.com> Co-authored-by: Vidushi Gupta <vidushig@nvidia.com> Co-authored-by: sarath-nalluri <72062105+sarath-nalluri@users.noreply.github.com> Co-authored-by: agentic-bug-fix <agentic-bug-fix@local> Co-authored-by: smasurekar <smasurekar@nvidia.com> Co-authored-by: nvskills-svc-account <svc-nvskills-signing@nvidia.com> * docs: highlight Agentic RAG in top-level README (#672) Overview, Key Features (Agentic RAG and Search and Retrieval), and documentation links for release-v2.6.0. (cherry picked from commit e69a8d1b5ea1bc469b360d59cf00d5ff3c89e555) Signed-off-by: Kurt Heiss <kheiss@nvidia.com> Co-authored-by: Kurt Heiss <kheiss@nvidia.com> * chore: polish RAG 2.6 sanity feedback notes (#670) * chore: polish launchable feedback notes Signed-off-by: Shubhadeep Das <shubhadeepd@nvidia.com> * fix: import fpdf positioning enums correctly Signed-off-by: Shubhadeep Das <shubhadeepd@nvidia.com> --------- Signed-off-by: Shubhadeep Das <shubhadeepd@nvidia.com> * Update docs/vlm.md Co-authored-by: coderabbitai[bot] <136622811+coderabbitai[bot]@users.noreply.github.com> * docs: restore main release notes in develop port Signed-off-by: Shubhadeep Das <shubhadeepd@nvidia.com> * fix: restore missed main port content Signed-off-by: Shubhadeep Das <shubhadeepd@nvidia.com> --------- Signed-off-by: Swapnil Masurekar <smasurekar@nvidia.com> Signed-off-by: Niyati Singal <nsingal@nvidia.com> Signed-off-by: Punit Kumar <punitk@nvidia.com> Signed-off-by: Minh Nguyen <minhngu@nvidia.com> Signed-off-by: smasurekar <smasurekar@nvidia.com> Signed-off-by: Sebastion <sebastiondev@users.noreply.github.com> Signed-off-by: Raj Sahoo <raj.sahoo@netapp.com> Signed-off-by: Truong Nguyen <tgnguyen@nvidia.com> Signed-off-by: richa <ricsingh@nvidia.com> Signed-off-by: Phuc Nguyen <pdnguyen@nvidia.com> Signed-off-by: Vidushi Gupta <vidushig@nvidia.com> Signed-off-by: Pranjal Doshi <pranjald@nvidia.com> Signed-off-by: Shubhadeep Das <shubhadeepd@nvidia.com> Signed-off-by: Nikhil Kulkarni <nikkulkarni@nvidia.com> Signed-off-by: Kurt Heiss <kheiss@nvidia.com> Signed-off-by: shubhadeepd <shubhadeepd@local> Signed-off-by: Richa Singh <ricsingh@nvidia.com> Signed-off-by: agentic-bug-fix <agentic-bug-fix@local> Signed-off-by: nvskills-svc-account <svc-nvskills-signing@nvidia.com> Co-authored-by: Swapnil Masurekar <smasurekar@nvidia.com> Co-authored-by: niyatisingal <nsingal@nvidia.com> Co-authored-by: kumar-punit <punitk@nvidia.com> Co-authored-by: Kurt Heiss <kheiss@nvidia.com> Co-authored-by: rkharwar-nv <rkharwar@nvidia.com> Co-authored-by: nkmcalli <nkmcalli@yahoo.com> Co-authored-by: Pranjal Doshi <pranjald@nvidia.com> Co-authored-by: nv-pranjald <150428320+nv-pranjald@users.noreply.github.com> Co-authored-by: Minh Nguyen <minhngu@nvidia.com> Co-authored-by: anngu <anngu@nvidia.com> Co-authored-by: Cursor <cursoragent@cursor.com> Co-authored-by: Nikhil Kulkarni <nikkulkarni@nvidia.com> Co-authored-by: nkmcalli <nmcallister@nvidia.com> Co-authored-by: Sebastion <sebastion@sebastion.dev> Co-authored-by: sahoor-netapp <raj.sahoo@netapp.com> Co-authored-by: Truong Nguyen <tgnguyen@nvidia.com> Co-authored-by: vidushig-nv <vidushig@nvidia.com> Co-authored-by: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com> Co-authored-by: richa-nvidia <ricsingh@nvidia.com> Co-authored-by: pdnguyen-hub <pdnguyen@nvidia.com> Co-authored-by: shubhadeepd <shubhadeepd@local> Co-authored-by: sarath-nalluri <72062105+sarath-nalluri@users.noreply.github.com> Co-authored-by: agentic-bug-fix <agentic-bug-fix@local> Co-authored-by: nvskills-svc-account <svc-nvskills-signing@nvidia.com> Co-authored-by: coderabbitai[bot] <136622811+coderabbitai[bot]@users.noreply.github.com>

smasurekar requested review from kumar-punit, niyatisingal, nv-nikkulkarni, nv-pranjald and shubhadeepd February 19, 2026 15:10

smasurekar added the enhancement New feature or request label Feb 19, 2026

smasurekar force-pushed the dev/smasurekar/nemotron-nims branch 4 times, most recently from 6d8876e to 07e98c9 Compare February 24, 2026 07:14

shubhadeepd added the release-26.03 label Feb 26, 2026

smasurekar force-pushed the dev/smasurekar/nemotron-nims branch from 07e98c9 to decfdd1 Compare February 27, 2026 06:19

Upgrade to Nemotron RC NIMs and NV-Ingest 26.1.2

cd8e9a3

Signed-off-by: Swapnil Masurekar <smasurekar@nvidia.com>

smasurekar force-pushed the dev/smasurekar/nemotron-nims branch from decfdd1 to cd8e9a3 Compare February 27, 2026 08:55

shubhadeepd approved these changes Feb 27, 2026

View reviewed changes

shubhadeepd merged commit 7d0796d into release-v2.5.0 Feb 27, 2026
6 checks passed

shubhadeepd deleted the dev/smasurekar/nemotron-nims branch February 27, 2026 10:16

shubhadeepd pushed a commit that referenced this pull request Mar 8, 2026

Upgrade to Nemotron RC NIMs and NV-Ingest 26.1.2 (#371)

0e6f3e4

Signed-off-by: Swapnil Masurekar <smasurekar@nvidia.com>

Navigation Menu

Uh oh!

Upgrade to Nemotron RC NIMs and NV-Ingest 26.1.2 - #371

Upgrade to Nemotron RC NIMs and NV-Ingest 26.1.2#371
shubhadeepd merged 1 commit into
release-v2.5.0from
dev/smasurekar/nemotron-nims

smasurekar commented Feb 19, 2026 •

edited

Loading

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Uh oh!

Conversation

smasurekar commented Feb 19, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Upgrade to Nemotron RC NIMs and NV-Ingest 26.1.2

Summary

NIM / model changes

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

smasurekar commented Feb 19, 2026 •

edited

Loading