# Nvidia <> Jan 26 Jan

## Products

- Jan Desktop
- Professional-grade GPUs
- Jan Server Suite
- Datacenter-grade GPUs
- H Series: Tensor Compute optimizations
- Nvidia Containers
- https://docs.nvidia.com/nemo-framework/user-guide/latest/llama/deploy.html

## Strategy

- Demos
- llama.cpp
- Production-grade
- A100 or H100, DGX Cloud
- TensorRT
- Use it for speed, optimizations
- https://developer.nvidia.com/blog/nvidia-tensorrt-llm-supercharges-large-language-model-inference-on-nvidia-h100-gpus/
- LLMOps Team
- Build and Tune a RAG
- Observability
- Anything that removes 10-20% of spend

## TensorRT & Triton Roadmap

- Interweaved
- Performance Improvements:
- https://nvidia.github.io/TensorRT-LLM/performance.html
- Triton will Suppport multiple backends
- TensorRT
- vLLM
- llama.cpp in Pytorch
- Model Deployment
- https://docs.nvidia.com/nemo-framework/user-guide/latest/llama/deploy.html
- Support for top models
- Mistral support

## Working with Nvidia

- Inception
- Strategic Support
- Should we upgrade with the H100?
- NVAIE = Enterprise Support
- File tickets thru NVAIE
- Prod-grade support

## How do we sell Nvidia?

- If clients/partners need to buy Nvidia?
- GTM with Nvidia?
- Angle 1: TensorRT Adoption
- Loop Ben in (Eng) for Professional, Enterprise-grade
- Demo with TensorRT?
- Release a blog (Nvidia Inception)
- NeMo
- Boutique SIs
- Partner Kits (how to position Nvidia)
- Contact Nvidia Resellers?
- GTM with Nvidia
- Cloud Validation
- App Catalog
- Inception
- AWS, GCP, Azure = Clouds
- Oracle Cloud
- https://www.nvidia.com/en-us/gpu-accelerated-applications/?search=SDK
- Development Journeys
- RTX Cards = Pending
- Workstream 1: Benchmnark for TensorRT-LLM with Triton (Jan Server)
- Alpha version: ASAP = help a lot
- Blog Post, etc
- Workstream 2: Jan's GTM
- Seldon Core? = MLOps
- Enterprise Support, Self-serve
- SDK with Public/Enterprise Offering
- Self-serve will have different identity from
- Certified with clouds
- Docker containers certified for Infosec
- Professional Services GTM
- Hiring

## Strategic Partners

- Investments
- June 2024 as a Raise window