ckg-nvidia-ai 0.1.0__tar.gz

ckg_nvidia_ai-0.1.0/.gitignore

@@ -0,0 +1,8 @@
+__pycache__/
+*.py[cod]
+*.egg-info/
+dist/
+build/
+.venv/
+venv/
+.env

ckg_nvidia_ai-0.1.0/PKG-INFO

@@ -0,0 +1,188 @@
+Metadata-Version: 2.4
+Name: ckg-nvidia-ai
+Version: 0.1.0
+Summary: NVIDIA AI developer stack as a traversable knowledge graph — 20 CKGs, 998 nodes, MCP-native
+Project-URL: Homepage, https://graphifymd.com
+Project-URL: Repository, https://github.com/Yarmoluk/ckg-nvidia-ai
+Project-URL: Bug Tracker, https://github.com/Yarmoluk/ckg-nvidia-ai/issues
+Author-email: Daniel Yarmoluk <daniel.yarmoluk@gmail.com>
+License: MIT
+Keywords: ai-agents,cuda,knowledge-graph,llm,mcp,nemo,nim,nvidia,tensorrt
+Classifier: Development Status :: 4 - Beta
+Classifier: Intended Audience :: Developers
+Classifier: License :: OSI Approved :: MIT License
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
+Requires-Python: >=3.11
+Requires-Dist: mcp[cli]>=1.0.0
+Description-Content-Type: text/markdown
+
+# ckg-nvidia-ai
+
+**The NVIDIA AI developer stack as a traversable knowledge graph.**
+
+20 CKGs · 998 nodes · MCP-native · 4× F1 of RAG · 11× fewer tokens · auditable by design
+
+---
+
+Instead of sending an AI agent to scan thousands of pages of NVIDIA documentation, give it a graph it can traverse. Every concept, every dependency, every connection — declared, typed, and queryable in ~269 tokens per question.
+
+```
+NIM → TensorRT-LLM → quantization → FP8 precision → Hopper SM90 requirement
+```
+
+The agent calls `query_ckg()` and gets that chain. Not a summary. The actual dependency path.
+
+---
+
+## What's inside
+
+| Domain | Description |
+|---|---|
+| `nvidia-nim` | NVIDIA Inference Microservices — deployment, scaling, speculative decoding |
+| `nvidia-nemo` | NeMo framework — training, PEFT, guardrails, evaluation |
+| `nvidia-tensorrt-triton` | TensorRT-LLM + Triton Inference Server — quantization, batching, KV cache |
+| `nvidia-cuda-toolkit` | CUDA compiler, PTX, memory hierarchy, Hopper/Blackwell features |
+| `nvidia-cuda-x-libraries` | cuBLAS, cuDNN, cuFFT, NCCL, Thrust — the acceleration layer |
+| `nvidia-hpc-sdk` | OpenACC, OpenMP, CUDA Fortran, multi-GPU scaling |
+| `nvidia-omniverse` | Universal Scene Description, simulation, digital twins |
+| `nvidia-isaac` | Isaac Lab + Isaac Sim — robot learning, sensor simulation |
+| `nvidia-cosmos` | Physical AI world foundation models — video generation, tokenization |
+| `nvidia-drive` | Autonomous vehicle stack — perception, planning, safety validation |
+| `nvidia-jetson` | Edge AI platform — Orin NX, AGX, DeepStream, Holoscan |
+| `nvidia-clara` | Healthcare AI — MONAI, Parabricks genomics, BioNeMo, Holoscan SDK |
+| `nvidia-metropolis` | Intelligent video analytics — VLMs, TAO Toolkit, DeepStream |
+| `nvidia-riva` | Speech AI — ASR, TTS, NLP pipelines, streaming |
+| `nvidia-gameworks` | Graphics R&D — DLSS, RTX, PhysX, Reflex |
+| `nvidia-developer-tools` | Nsight, CUPTI, Compute Sanitizer, profiling stack |
+| `nvidia-graphics-research` | Research graphics — neural rendering, path tracing, differentiable rendering |
+| `nvidia-ai-enterprise` | Enterprise AI platform — NIM blueprints, governance, fleet management |
+| `nvidia-developer-ecosystem` | Cross-cutting: NGC, DGX, Inception, AgentIQ, MCP integration |
+| `nvidia-openshell` | Agent sandbox runtime — policy enforcement, CVEs, authorization gaps |
+
+---
+
+## Install
+
+```bash
+pip install ckg-nvidia-ai
+```
+
+Or run without installing:
+
+```bash
+uvx ckg-nvidia-ai
+```
+
+---
+
+## Use as MCP Server
+
+### Claude Desktop
+
+```json
+{
+  "mcpServers": {
+    "nvidia-ai": {
+      "command": "uvx",
+      "args": ["ckg-nvidia-ai"]
+    }
+  }
+}
+```
+
+### Cursor / other MCP clients
+
+Same config — substitute `uvx` with `python -m ckg_nvidia_ai` if you prefer a venv install.
+
+---
+
+## Tools
+
+### `list_domains()`
+Returns all 20 NVIDIA AI domains. Start here.
+
+### `search_concepts(query, domain)`
+Find concepts by keyword within a domain.
+
+```
+search_concepts("speculative decoding", "nvidia-nim")
+→ Speculative Decoding [Optimization]
+   Draft Model [Component]
+   KV Cache [Infrastructure]
+```
+
+### `query_ckg(concept, domain, depth=3)`
+Traverse the graph from a concept — see what it requires and what depends on it.
+
+```
+query_ckg("TensorRT-LLM", "nvidia-tensorrt-triton", 3)
+→ ## TensorRT-LLM · nvidia-tensorrt-triton
+   Type: Framework
+
+   ### Prerequisites (what you need first)
+     - CUDA Toolkit
+       - CUDA Driver API
+       - cuBLAS
+     - Hopper SM90 Architecture
+     - FP8 / FP4 Quantization
+
+   ### Builds toward
+     - Triton Inference Server
+     - NIM Microservice Runtime
+```
+
+### `get_prerequisites(concept, domain)`
+Full ordered prerequisite chain — everything to understand or install first.
+
+```
+get_prerequisites("Isaac Lab", "nvidia-isaac")
+→ Isaac Lab → Isaac Sim → USD Composer → Omniverse Kit → ...
+```
+
+---
+
+## How it works
+
+Each domain is a typed dependency graph stored as CSV:
+
+```
+ConceptID, ConceptLabel, Dependencies, TaxonomyID
+1, TensorRT-LLM, "", Framework
+2, CUDA Toolkit, "", Platform
+3, FP8 Quantization, "2", Optimization
+4, Hopper SM90, "2", Architecture
+5, Speculative Decoding, "1|4", Optimization
+```
+
+When an agent queries a concept, the server runs BFS traversal over declared edges. The answer is composed entirely of traversed relationships — not probabilistic inference, not RAG retrieval, not token prediction over documentation.
+
+**The graph doesn't guess. It traverses.**
+
+---
+
+## Benchmark
+
+Built on the [KRB Benchmark v0.6.2](https://github.com/Yarmoluk/ckg-benchmark/blob/main/paper/main.pdf):
+
+| System | F1 | Tokens/query | Cost |
+|---|---|---|---|
+| **CKG** | **0.471** | **269** | **$7.81/1K** |
+| RAG | 0.123 | 2,982 | $76.23/1K |
+| GraphRAG | 0.120 | — | — |
+
+~4× F1 · 11× fewer tokens · auditable by design
+
+---
+
+## Related
+
+- **[ckg-mcp](https://pypi.org/project/ckg-mcp/)** — 97 domains across all topics (includes NVIDIA + science, finance, law, healthcare, and more)
+- **[KRB Benchmark](https://huggingface.co/datasets/danyarm/ckg-benchmark)** — open benchmark dataset
+- **[graphifymd.com](https://graphifymd.com)** — CKG catalog and Context-as-a-Service
+
+---
+
+*Built by [Graphify.md](https://graphifymd.com). Patent pending.*

ckg_nvidia_ai-0.1.0/README.md

@@ -0,0 +1,167 @@
+# ckg-nvidia-ai
+
+**The NVIDIA AI developer stack as a traversable knowledge graph.**
+
+20 CKGs · 998 nodes · MCP-native · 4× F1 of RAG · 11× fewer tokens · auditable by design
+
+---
+
+Instead of sending an AI agent to scan thousands of pages of NVIDIA documentation, give it a graph it can traverse. Every concept, every dependency, every connection — declared, typed, and queryable in ~269 tokens per question.
+
+```
+NIM → TensorRT-LLM → quantization → FP8 precision → Hopper SM90 requirement
+```
+
+The agent calls `query_ckg()` and gets that chain. Not a summary. The actual dependency path.
+
+---
+
+## What's inside
+
+| Domain | Description |
+|---|---|
+| `nvidia-nim` | NVIDIA Inference Microservices — deployment, scaling, speculative decoding |
+| `nvidia-nemo` | NeMo framework — training, PEFT, guardrails, evaluation |
+| `nvidia-tensorrt-triton` | TensorRT-LLM + Triton Inference Server — quantization, batching, KV cache |
+| `nvidia-cuda-toolkit` | CUDA compiler, PTX, memory hierarchy, Hopper/Blackwell features |
+| `nvidia-cuda-x-libraries` | cuBLAS, cuDNN, cuFFT, NCCL, Thrust — the acceleration layer |
+| `nvidia-hpc-sdk` | OpenACC, OpenMP, CUDA Fortran, multi-GPU scaling |
+| `nvidia-omniverse` | Universal Scene Description, simulation, digital twins |
+| `nvidia-isaac` | Isaac Lab + Isaac Sim — robot learning, sensor simulation |
+| `nvidia-cosmos` | Physical AI world foundation models — video generation, tokenization |
+| `nvidia-drive` | Autonomous vehicle stack — perception, planning, safety validation |
+| `nvidia-jetson` | Edge AI platform — Orin NX, AGX, DeepStream, Holoscan |
+| `nvidia-clara` | Healthcare AI — MONAI, Parabricks genomics, BioNeMo, Holoscan SDK |
+| `nvidia-metropolis` | Intelligent video analytics — VLMs, TAO Toolkit, DeepStream |
+| `nvidia-riva` | Speech AI — ASR, TTS, NLP pipelines, streaming |
+| `nvidia-gameworks` | Graphics R&D — DLSS, RTX, PhysX, Reflex |
+| `nvidia-developer-tools` | Nsight, CUPTI, Compute Sanitizer, profiling stack |
+| `nvidia-graphics-research` | Research graphics — neural rendering, path tracing, differentiable rendering |
+| `nvidia-ai-enterprise` | Enterprise AI platform — NIM blueprints, governance, fleet management |
+| `nvidia-developer-ecosystem` | Cross-cutting: NGC, DGX, Inception, AgentIQ, MCP integration |
+| `nvidia-openshell` | Agent sandbox runtime — policy enforcement, CVEs, authorization gaps |
+
+---
+
+## Install
+
+```bash
+pip install ckg-nvidia-ai
+```
+
+Or run without installing:
+
+```bash
+uvx ckg-nvidia-ai
+```
+
+---
+
+## Use as MCP Server
+
+### Claude Desktop
+
+```json
+{
+  "mcpServers": {
+    "nvidia-ai": {
+      "command": "uvx",
+      "args": ["ckg-nvidia-ai"]
+    }
+  }
+}
+```
+
+### Cursor / other MCP clients
+
+Same config — substitute `uvx` with `python -m ckg_nvidia_ai` if you prefer a venv install.
+
+---
+
+## Tools
+
+### `list_domains()`
+Returns all 20 NVIDIA AI domains. Start here.
+
+### `search_concepts(query, domain)`
+Find concepts by keyword within a domain.
+
+```
+search_concepts("speculative decoding", "nvidia-nim")
+→ Speculative Decoding [Optimization]
+   Draft Model [Component]
+   KV Cache [Infrastructure]
+```
+
+### `query_ckg(concept, domain, depth=3)`
+Traverse the graph from a concept — see what it requires and what depends on it.
+
+```
+query_ckg("TensorRT-LLM", "nvidia-tensorrt-triton", 3)
+→ ## TensorRT-LLM · nvidia-tensorrt-triton
+   Type: Framework
+
+   ### Prerequisites (what you need first)
+     - CUDA Toolkit
+       - CUDA Driver API
+       - cuBLAS
+     - Hopper SM90 Architecture
+     - FP8 / FP4 Quantization
+
+   ### Builds toward
+     - Triton Inference Server
+     - NIM Microservice Runtime
+```
+
+### `get_prerequisites(concept, domain)`
+Full ordered prerequisite chain — everything to understand or install first.
+
+```
+get_prerequisites("Isaac Lab", "nvidia-isaac")
+→ Isaac Lab → Isaac Sim → USD Composer → Omniverse Kit → ...
+```
+
+---
+
+## How it works
+
+Each domain is a typed dependency graph stored as CSV:
+
+```
+ConceptID, ConceptLabel, Dependencies, TaxonomyID
+1, TensorRT-LLM, "", Framework
+2, CUDA Toolkit, "", Platform
+3, FP8 Quantization, "2", Optimization
+4, Hopper SM90, "2", Architecture
+5, Speculative Decoding, "1|4", Optimization
+```
+
+When an agent queries a concept, the server runs BFS traversal over declared edges. The answer is composed entirely of traversed relationships — not probabilistic inference, not RAG retrieval, not token prediction over documentation.
+
+**The graph doesn't guess. It traverses.**
+
+---
+
+## Benchmark
+
+Built on the [KRB Benchmark v0.6.2](https://github.com/Yarmoluk/ckg-benchmark/blob/main/paper/main.pdf):
+
+| System | F1 | Tokens/query | Cost |
+|---|---|---|---|
+| **CKG** | **0.471** | **269** | **$7.81/1K** |
+| RAG | 0.123 | 2,982 | $76.23/1K |
+| GraphRAG | 0.120 | — | — |
+
+~4× F1 · 11× fewer tokens · auditable by design
+
+---
+
+## Related
+
+- **[ckg-mcp](https://pypi.org/project/ckg-mcp/)** — 97 domains across all topics (includes NVIDIA + science, finance, law, healthcare, and more)
+- **[KRB Benchmark](https://huggingface.co/datasets/danyarm/ckg-benchmark)** — open benchmark dataset
+- **[graphifymd.com](https://graphifymd.com)** — CKG catalog and Context-as-a-Service
+
+---
+
+*Built by [Graphify.md](https://graphifymd.com). Patent pending.*

ckg_nvidia_ai-0.1.0/pyproject.toml

@@ -0,0 +1,34 @@
+[build-system]
+requires = ["hatchling"]
+build-backend = "hatchling.build"
+
+[project]
+name = "ckg-nvidia-ai"
+version = "0.1.0"
+description = "NVIDIA AI developer stack as a traversable knowledge graph — 20 CKGs, 998 nodes, MCP-native"
+readme = "README.md"
+license = { text = "MIT" }
+authors = [{ name = "Daniel Yarmoluk", email = "daniel.yarmoluk@gmail.com" }]
+keywords = ["nvidia", "knowledge-graph", "mcp", "ai-agents", "nim", "nemo", "cuda", "tensorrt", "llm"]
+classifiers = [
+    "Development Status :: 4 - Beta",
+    "Intended Audience :: Developers",
+    "License :: OSI Approved :: MIT License",
+    "Programming Language :: Python :: 3",
+    "Programming Language :: Python :: 3.11",
+    "Programming Language :: Python :: 3.12",
+    "Topic :: Scientific/Engineering :: Artificial Intelligence",
+]
+requires-python = ">=3.11"
+dependencies = ["mcp[cli]>=1.0.0"]
+
+[project.scripts]
+ckg-nvidia-ai = "ckg_nvidia_ai.server:main"
+
+[project.urls]
+Homepage = "https://graphifymd.com"
+Repository = "https://github.com/Yarmoluk/ckg-nvidia-ai"
+"Bug Tracker" = "https://github.com/Yarmoluk/ckg-nvidia-ai/issues"
+
+[tool.hatch.build.targets.wheel]
+packages = ["src/ckg_nvidia_ai"]

ckg_nvidia_ai-0.1.0/src/ckg_nvidia_ai/__init__.py

@@ -0,0 +1,2 @@
+"""ckg-nvidia-ai — NVIDIA AI developer stack as a traversable knowledge graph."""
+__version__ = "0.1.0"

ckg_nvidia_ai-0.1.0/src/ckg_nvidia_ai/__main__.py

@@ -0,0 +1,2 @@
+from .server import main
+main()

ckg_nvidia_ai-0.1.0/src/ckg_nvidia_ai/domains/nvidia-ai-enterprise.csv

@@ -0,0 +1,51 @@
+ConceptID,ConceptLabel,Dependencies,TaxonomyID
+1,NVIDIA AI Enterprise,2|3,Platform
+2,NVAIE License,,Commercial
+3,Enterprise Support,1,Support
+4,Long-Term Support Branch,1,Policy
+5,90-Day Trial,,Commercial
+6,Subscription Tiers,2,Commercial
+7,License Renewal,2,Commercial
+8,CVE Patching SLA,3,Policy
+9,Compliance Frameworks,8,Regulatory
+10,Enterprise Support Portal,3,Tool
+11,License Server,2,Infrastructure
+12,NGC Private Registry,2,Registry
+13,AI Enterprise Hub,12,Catalog
+14,Hardened Container Images,12|2,Artifact
+15,Container Security Scanning,14,Security
+16,Certified Software Catalog,13,Catalog
+17,NVIDIA API Catalog,13|2,Service
+18,NIM Enterprise,2|21,Microservice
+19,NeMo Enterprise,2|23,Framework
+20,RAPIDS Enterprise,2,Library
+21,Triton Enterprise,2|23,Server
+22,TensorRT Enterprise,2|23,Library
+23,CUDA Enterprise Runtime,2,Runtime
+24,cuOpt Enterprise,2|23,Microservice
+25,Merlin Enterprise,2|20,Framework
+26,Parabricks Enterprise,2|23,Application
+27,DeepStream Enterprise,2|22,SDK
+28,Riva Enterprise,2|21,Microservice
+29,MONAI Enterprise,2,Framework
+30,GPU Operator,33,Operator
+31,Network Operator,33,Operator
+32,NIM Operator,33,Operator
+33,Kubernetes Cluster,,Infrastructure
+34,Run:AI Integration,33,Integration
+35,DCGM,30,Monitor
+36,Prometheus / Grafana Stack,35,Monitor
+37,DOCA Enterprise,31,SDK
+38,Cloud Partners,,Partner
+39,DGX Cloud,40,Service
+40,Base Command Platform,,Platform
+41,Certified Systems,2,Hardware
+42,Private Cloud,41,Infrastructure
+43,vGPU Software,2,Software
+44,Multi-Node Training,40,Workload
+45,Fleet Command,2,Platform
+46,Jetson Edge Fleet,45,Hardware
+47,ISV Partner Program,2,Program
+48,MLOps Pipeline,19|20,Workflow
+49,AI Workflow Catalog,13,Catalog
+50,NVAIE Manager Portal,1,Tool

ckg_nvidia_ai-0.1.0/src/ckg_nvidia_ai/domains/nvidia-clara.csv

@@ -0,0 +1,46 @@
+ConceptID,ConceptLabel,Dependencies,TaxonomyID
+1,Clara Platform,,Platform
+2,Cuda X Libraries,,SDK
+3,Nvidia Ngc,,Platform
+4,Dgx Systems,,Platform
+5,Base Command Platform,4,Platform
+6,Bionemo Agent Toolkit,2,Platform
+7,Agent Skills Framework,6,SDK
+8,Skill Cards,7,Standard
+9,Digital Biology Skills,7,Concept
+10,Digital Health Skills,7,Concept
+11,Medtech Skills,7,Concept
+12,Closed Loop Discovery,6,Concept
+13,Parabricks,2,SDK
+14,Gatk4 Best Practices,,Standard
+15,Variant Calling,14,Workflow
+16,Short Read Sequencing,13,Workflow
+17,Long Read Sequencing,13,Workflow
+18,Gpu Seq Acceleration,2,Algorithm
+19,Monai Toolkit,2|3,SDK
+20,Monai Core,,SDK
+21,Monai Label,20,Tool
+22,Nvidia Flare,,SDK
+23,Federated Learning,22,Algorithm
+24,Clara Viz,2,SDK
+25,Monai Pretrained Models,19,Concept
+26,Bionemo Training Framework,2|4|5,SDK
+27,Protein Structure Prediction,26,Algorithm
+28,Molecular Docking,26,Algorithm
+29,Generative Chemistry,26,Algorithm
+30,Generative Protein Design,26,Algorithm
+31,Protein Binder Design,27|28|30,Workflow
+32,Biomolecular Formats,,Standard
+33,Rapids Singlecell,2,SDK
+34,Single Cell Transcriptomics,33,Workflow
+35,Dimensionality Reduction Sc,33,Algorithm
+36,Gpu Clustering,33,Algorithm
+37,Holoscan Sdk,2,SDK
+38,Isaac Healthcare,2,Platform
+39,Surgical Robotics Policy,38,Workflow
+40,Medical Device Streaming,37,Workflow
+41,Biomedical Aiq Agent,6,Tool
+42,Ambient Healthcare Agents,6,Tool
+43,Clinical Data Privacy,23,Concept
+44,Automated Documentation,42,Workflow
+45,Real World Evidence,,Concept

ckg_nvidia_ai-0.1.0/src/ckg_nvidia_ai/domains/nvidia-cosmos.csv

@@ -0,0 +1,47 @@
+ConceptID,ConceptLabel,Dependencies,TaxonomyID
+1,Cosmos World Foundation Model,24|29|21,PLATFORM
+2,Cosmos-1.0-Predict,1|4|23|22,MODEL
+3,Cosmos-1.0-Transfer,1|17,MODEL
+4,Cosmos Tokenizer,10|31|32,TOOL
+5,Cosmos-1.0-Reason,1,MODEL
+6,Cosmos-1.0-Nano,1|36,MODEL
+7,Physical AI,15|16,CONCEPT
+8,Synthetic Data Pipeline,2|45,PIPELINE
+9,Sim-to-Real Transfer,3|43,CONCEPT
+10,Video Tokenization,4,PROCESS
+11,Autoregressive WFM,32,ARCHITECTURE
+12,Diffusion WFM,31|30,ARCHITECTURE
+13,Text-to-World Generation,2|40,CAPABILITY
+14,Video-to-World Prediction,2|10,CAPABILITY
+15,Robot Policy Training,8|23|43,PROCESS
+16,AV Scenario Simulation,2|13|22,PROCESS
+17,Omniverse Integration,45,INTEGRATION
+18,Isaac Lab Integration,15,INTEGRATION
+19,DRIVE Sim Integration,16,INTEGRATION
+20,GR00T Integration,18|15,INTEGRATION
+21,Physical Laws Conditioning,42,MECHANISM
+22,Camera Model Conditioning,40,MECHANISM
+23,Action Conditioning,40,MECHANISM
+24,Cosmos Dataset,35,DATA
+25,Safety Filter,,COMPONENT
+26,NVIDIA Open Model License,,POLICY
+27,HuggingFace Release,26,DISTRIBUTION
+28,Post-Training / Fine-Tuning,37|4,PROCESS
+29,Spatiotemporal Attention,,MECHANISM
+30,Video Diffusion Transformer (DiT),29,ARCHITECTURE
+31,Continuous Tokenizer,,COMPONENT
+32,Discrete Tokenizer,,COMPONENT
+33,World State Representation,10,CONCEPT
+34,Multi-View Generation,22|2,CAPABILITY
+35,Teleoperation Data,,DATA
+36,Edge Deployment,,PROCESS
+37,NeMo Framework,,TOOL
+38,WFM Inference API,,API
+39,Generative World Simulation,1,CONCEPT
+40,Conditioning Framework,,MECHANISM
+41,Video Pretraining Corpus,24,DATA
+42,Physics Simulation Engine,17,TOOL
+43,Domain Randomization,9,TECHNIQUE
+44,Robot Manipulation,15,CAPABILITY
+45,3D Scene Generation,17,CAPABILITY
+46,NVIDIA Cloud Functions (NVCF),,PLATFORM

ckg_nvidia_ai-0.1.0/src/ckg_nvidia_ai/domains/nvidia-cuda-toolkit.csv

@@ -0,0 +1,51 @@
+ConceptID,ConceptLabel,Dependencies,TaxonomyID
+1,Nvcc Compiler,2|3|4|7|8,Tool
+2,Ptx Ir,3|6,Standard
+3,Cubin,,Tool
+4,Fatbinary,2|3,Tool
+5,Nvrtc,2|6,SDK
+6,Ptx Jit,3,Algorithm
+7,Libdevice,9,SDK
+8,Cuda Cpp Extensions,11|17|12|33,Standard
+9,Cuda Math Api,,API
+10,Thread Hierarchy,11|12|13,Concept
+11,Warp,,Concept
+12,Cooperative Groups,11,API
+13,Cuda Cluster,44,Concept
+14,Occupancy Calculator,16,Tool
+15,Sm Architecture,,Concept
+16,Device Properties,,API
+17,Shared Memory,,Concept
+18,Global Memory,19,Concept
+19,Hbm Memory,,Platform
+20,L2 Cache,,Concept
+21,Unified Memory,18|29,API
+22,Pinned Memory,,Concept
+23,Memory Pool,18,API
+24,Texture Memory,,Concept
+25,Constant Memory,,Concept
+26,Cudamemcpyasync,30|22,API
+27,Peer Access,49,API
+28,Cuda Runtime,29|30|31|23,API
+29,Cuda Driver Api,6|15,API
+30,Stream Api,,API
+31,Event Api,30,API
+32,Cuda Graphs,30|28,API
+33,Atomic Operations,,Concept
+34,Cublas,28|9,SDK
+35,Cusparse,28,SDK
+36,Cudss,35|28,SDK
+37,Thrust,38|28,SDK
+38,Cub,15,SDK
+39,Cupti,28,SDK
+40,Nsight Systems,39|30,Tool
+41,Nsight Compute,39|14,Tool
+42,Compute Sanitizer,28|17,Tool
+43,Hopper Architecture,44|45|13|49,Platform
+44,Tma Hopper,17|18,API
+45,Warpgroup Mma,17|11,API
+46,Blackwell Architecture,47|48|49,Platform
+47,Fp4 Blackwell,45,Concept
+48,Mxfp8,45|34,Concept
+49,Nvlink,,Platform
+50,Multi Gpu Api,27|28|49,API

ckg_nvidia_ai-0.1.0/src/ckg_nvidia_ai/domains/nvidia-cuda-x-libraries.csv

@@ -0,0 +1,51 @@
+ConceptID,ConceptLabel,Dependencies,TaxonomyID
+1,Cuda Foundation,,Platform
+2,Tensor Cores,1,Platform
+3,Cublas,1|2,API
+4,Cufft,1,API
+5,Curand,1,API
+6,Cusolver,1|3,API
+7,Cusparse,1,API
+8,Cutensor,1|2,API
+9,Cudss,1|6,API
+10,Amgx,1|7|6,API
+11,Nvmath Python,3|4,SDK
+12,Cudnn,1|2|3,API
+13,Operation Graph,12,Concept
+14,Multi Op Fusion,13,Algorithm
+15,Frontend Api,12|13,API
+16,Cutlass,1|2,SDK
+17,Flashinfer,16|2,API
+18,Cuda Graph,1,Tool
+19,Tensorrt,12|3|16,SDK
+20,Tensorrt Llm,19|17|39|41,SDK
+21,Thrust,1,API
+22,Cub,1,API
+23,Cuda Parallel,21|22,API
+24,Rapids,1|21,Platform
+25,Cudf,24,API
+26,Cuml,24|3,API
+27,Cugraph,24|7,API
+28,Cuvs,24,API
+29,Cuopt,1|3,API
+30,Nemo Curator,25|28,Tool
+31,Morpheus,19|25|26,SDK
+32,Gpu Direct Storage,1,Platform
+33,Dali,1|34,SDK
+34,Nvimagecodec,1,API
+35,Cv Cuda,1|36,API
+36,Npp,1,API
+37,Video Codec Sdk,1,SDK
+38,Optical Flow Sdk,1,SDK
+39,Nccl,1,API
+40,Nvshmem,1|39,API
+41,Nixl,1|32,API
+42,Cuequivariance,1,API
+43,Culitho,1|4|7,API
+44,Warp Nvidia,1,SDK
+45,Physics Nemo,1|12|44|42,SDK
+46,Earth2,45|39|12,Platform
+47,Cuquantum,1|8,SDK
+48,Cupqc,1|5,API
+49,Cuda Q Qec,47,SDK
+50,Cuda Q Solvers,47|3,SDK